[R] need some lattice help
Hello everybody,
I just started with lattice plots and I was wondering why it behaves different
than expected.
If I generated multiple plots in the past, I just used some code like this:
lapply(..., {
windows()
plot(...)
})
I got multiple windows, each containing one plot.
Now, I tried the same with levelplot (instead of plot) and I see all
windows are created but the plots are only drawn in one window, one on top of
the other...
Can anybody give me a hint, why it behaves like this and what should I change?
Ciao,
Antje
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] Running R under Sun Grid Engine with OpenMPI tight integration
2009/1/27 Rainer M Krug r.m.k...@gmail.com: On Tue, Jan 27, 2009 at 2:30 AM, Peter Waltman peter.walt...@gmail.com wrote: Hi - I saw your posting on the R-help mailing list. Were you ever able to get this working? did you end up switching to use the rsge library? Yes - that is exactly what I did - I am using rsge or, which is in most cases sufficient for me, starting several instances of R and run the whole simulation (array processing). But I would still like to know how I can use the Rmpi and snow on the Sun Grid Engine. Please keep me posted, Rainer I'm trying to do the same, and not having very much luck getting it going. Thanks! Peter Waltman -- Rainer M. Krug, PhD (Conservation Ecology, SUN), MSc (Conservation Biology, UCT), Dipl. Phys. (Germany) Centre of Excellence for Invasion Biology Faculty of Science Natural Sciences Building Private Bag X1 University of Stellenbosch Matieland 7602 South Africa __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Hi all, I have been installed R in SGE but with LAM tight integration, I don't k now if this help you :s . The problem I had with OpenMPI was that the load in all nodes waiting for a job and with an instance of R. Their load up to 100 %, and there isn't a job running on them. Even so, I will try to configure it with OpenMPI..when I have results I will post them ok? Saludos __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Help
Hi, All, I want to construct a pareto chart for my data. My data has two columns: modules and defects(each module has several defects). Now I use pareto.chart to create a chart for defects which y axis is cumulative percentage of defect, and x axis is value of defect. That is not what I want. I need the x axis be percentage of modules. Is there any way that I can change the scale of x axis, that make x axis be percent scale for modules, like 0% to 100% modules. So that I can directly read the chart, is that 20% modules contains 80% defect. Or, is there any function that I can use to achive it directly? Thank you in advanced! Hedy __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Running R under Sun Grid Engine with OpenMPI tight integration
On Tue, Jan 27, 2009 at 11:04 AM, Marce marc...@gmail.com wrote: 2009/1/27 Rainer M Krug r.m.k...@gmail.com: On Tue, Jan 27, 2009 at 2:30 AM, Peter Waltman peter.walt...@gmail.com wrote: Hi - I saw your posting on the R-help mailing list. Were you ever able to get this working? did you end up switching to use the rsge library? Yes - that is exactly what I did - I am using rsge or, which is in most cases sufficient for me, starting several instances of R and run the whole simulation (array processing). But I would still like to know how I can use the Rmpi and snow on the Sun Grid Engine. Please keep me posted, Rainer I'm trying to do the same, and not having very much luck getting it going. Thanks! Peter Waltman -- Rainer M. Krug, PhD (Conservation Ecology, SUN), MSc (Conservation Biology, UCT), Dipl. Phys. (Germany) Centre of Excellence for Invasion Biology Faculty of Science Natural Sciences Building Private Bag X1 University of Stellenbosch Matieland 7602 South Africa __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Hi all, I have been installed R in SGE but with LAM tight integration, I don't k now if this help you :s . The problem I had with OpenMPI was that the load in all nodes waiting for a job and with an instance of R. Their load up to 100 %, and there isn't a job running on them. Even so, I will try to configure it with OpenMPI..when I have results I will post them ok? Yes please - could you please provide some info (and a small example) on how you managed this with LAM? I think we should move this thread to r-sig-hpc, as it is more appropriate there. I CC'd it to there - please respond there and not on r-help. Thanks Rainer Saludos __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Rainer M. Krug, PhD (Conservation Ecology, SUN), MSc (Conservation Biology, UCT), Dipl. Phys. (Germany) Centre of Excellence for Invasion Biology Faculty of Science Natural Sciences Building Private Bag X1 University of Stellenbosch Matieland 7602 South Africa __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] need some lattice help
I guess, I got the answer.
(http://cran.r-project.org/doc/FAQ/R-FAQ.html#Why-do-lattice_002ftrellis-graphics-not-work_003f)
Ciao,
Antje
Antje schrieb:
Hello everybody,
I just started with lattice plots and I was wondering why it behaves
different than expected.
If I generated multiple plots in the past, I just used some code like this:
lapply(..., {
windows()
plot(...)
})
I got multiple windows, each containing one plot.
Now, I tried the same with levelplot (instead of plot) and I see all
windows are created but the plots are only drawn in one window, one on
top of the other...
Can anybody give me a hint, why it behaves like this and what should I
change?
Ciao,
Antje
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
[R] Memory issue?
I have a script that sometimes produces the following error:
Error in assign(.target, met...@target, envir = envir) :
formal argument envir matched by multiple actual arguments
Do you think this is a memory issue? I don't know what else it could be
as it doesn't always occur even if the script is run with exactly the
same data.
Does rm() actually free up memory?
Thanks
Dan
--
**
Daniel Brewer, Ph.D.
Institute of Cancer Research
Molecular Carcinogenesis
Email: daniel.bre...@icr.ac.uk
**
The Institute of Cancer Research: Royal Cancer Hospital, a charitable Company
Limited by Guarantee, Registered in England under Company No. 534147 with its
Registered Office at 123 Old Brompton Road, London SW7 3RP.
This e-mail message is confidential and for use by the a...{{dropped:2}}
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
[R] Odp: working with tables -- was Re: Mode (statistics) in R?
Hi r-help-boun...@r-project.org napsal dne 27.01.2009 01:50:39: Ok, so I'm slowly figuring out what a factor is, and was able to follow the related thread about finding a mode by using constructs like my_mode = as.numeric(names(table(x))[which.max(table(x))]) Now, suppose I want to keep looking for other modes? For example, Do you mean the situation when there are more then one max values? set.seed(19) sample(seq(1,10),50,replace=TRUE)-bag table(bag)[max(table(bag))==table(bag)] Regards Petr Rgames sample(seq(1,10),50,replace=TRUE)-bag Rgames bag [1] 2 8 8 10 7 3 2 9 8 3 8 9 6 6 10 10 7 1 [19] 9 5 4 3 3 5 10 3 6 3 2 8 4 2 1 10 6 2 [37] 6 6 9 8 6 8 8 4 3 6 3 9 5 1 Rgames names(which.max(table(bag))) [1] 3 I can then do Rgames bag2-bag[bag!=3] and repeat the which.max stuff. I came up with the following command to find the actual magnitude of the mode: Rgames table(bag)-tbag Rgames tbag bag 1 2 3 4 5 6 7 8 9 10 3 5 8 3 3 8 2 8 5 5 Rgames tbag[dimnames(tbag)$bag==3]-bagmode Rgames bagmode 3 8 Related to this, since bag2 is now bereft of threes, Rgames table(bag2) bag2 1 2 4 5 6 7 8 9 10 3 5 3 3 8 2 8 5 5 I was able to make the same table with Rgames newtable-tbag[c(dimnames(tbag)$bag)!=3] Rgames newtable bag 1 2 4 5 6 7 8 9 10 3 5 3 3 8 2 8 5 5 Is there a cleaner syntax to do these things? Thanks for your help--and feel free to point me to the Inferno or other paper on the philosophy and use of factors and tables. Carl __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Need help on running Heckman Correction Estimation using R
Team, I am trying to resolve the self-selection bias of a sample in an experiment and would like to run the Heckman Correction Estimation using R. Can someone help me with the R-Code... I tried searching for the discussion, but not successful. Thanks in advance, Best, Kishore/.. http://kaykayatisb.blogspot.com [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Re : Need help on running Heckman Correction Estimation using R
See the micEcon package. there is and heckit function  Justin BEM BP 1917 Yaoundé Tél (237) 99597295 (237) 22040246 De : Kishore gladikish...@gmail.com à : r-help@r-project.org; r-h...@stat.math.ethz.ch Envoyé le : Mardi, 27 Janvier 2009, 11h54mn 00s Objet : [R] Need help on running Heckman Correction Estimation using R Team, I am trying to resolve the self-selection bias of a sample in an experiment and would like to run the Heckman Correction Estimation using R. Can someone help me with the R-Code... I tried searching for the discussion, but not successful. Thanks in advance, Best, Kishore/.. http://kaykayatisb.blogspot.com    [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Memory issue?
Daniel Brewer wrote: I have a script that sometimes produces the following error: Error in assign(.target, met...@target, envir = envir) : formal argument envir matched by multiple actual arguments Do you think this is a memory issue? I don't know what else it could be as it doesn't always occur even if the script is run with exactly the same data. Does rm() actually free up memory? Thanks Dan Hi, There are multiple threads on this subject on the R-help list, googling for formal argument matched by mutiple actual arguments lead me to: http://tolstoy.newcastle.edu.au/R/help/05/08/10698.html So this is probably not a memory issue. Freeing up memory can be done using gc(). cheers and hth, Paul -- Drs. Paul Hiemstra Department of Physical Geography Faculty of Geosciences University of Utrecht Heidelberglaan 2 P.O. Box 80.115 3508 TC Utrecht Phone: +31302535773 Fax:+31302531145 http://intamap.geo.uu.nl/~paul __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Goodness of fit for gamma distributions
Hi Dann, there is probably a better way to do this, but this works anyway: # your data gamdat - rgamma(1, shape=1, rate=0.5) # comparison to gamma: gamsam - rgamma(1, shape=1, rate=0.6) qqplot(gamsam,gamdat) abline(0,1) greetings Remko - Remko Duursma Post-Doctoral Fellow Centre for Plant and Food Science University of Western Sydney Hawkesbury Campus Richmond NSW 2753 Dept of Biological Science Macquarie University North Ryde NSW 2109 Australia Mobile: +61 (0)422 096908 On Tue, Jan 27, 2009 at 3:38 AM, Dan31415 d.m.mitch...@reading.ac.uk wrote: I'm looking for goodness of fit tests for gamma distributions with large data sizes. I have a matrix with around 10,000 data values in it and i have fitted a gamma distribution over a histogram of the data. The problem is testing how well that distribution fits. Chi-squared seems to be used more for discrete distributions and kolmogorov-smirnov seems that large sample sizes make it had to evaluate the D statistic. Also i haven't found a qq plot for gamma, although i think this might be an appropriate test. in summary -is there a gamma goodness of fit test that doesnt depend on the sample size? -is there a way of using qqplot for gamma distributions, if so how would you calculate it from a matrix of data values? regards, Dann -- View this message in context: http://www.nabble.com/Goodness-of-fit-for-gamma-distributions-tp21668711p21668711.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Error in Surv(time, status) : Time variable is not numeric
R is finding the R function 'time' rather than your variable time in your dataset. Perhaps adding 'data=mydata' to your function call will resolve your issue. Chris Braem M wrote: Dear, I want to analyze two-level survival data using a shared frailty model, for which I want to use the R package 'Frailtypack, proposed by Rondeau et al. The dataset was built using SAS software. I also tried to change the format using SPSS and Excell. My (reduced) dataset has following column names: ID entrytimestatusfamily var1 I used following command: frailtyPenal(Surv(time, status) ~var1 + cluster(family), Frailty=TRUE ,n.knots=8, kappa1=1500, + cross.validation=FALSE) And got this error : Error in Surv(time, status) : Time variable is not numeric In addition: Warning message: In is.na(time) : is.na() applied to non-(list or vector) of type 'closure' I think R transforms the data when importing into R, so that the observations are not numeric anymore. Does anyone know how to handle this problem? Thanks, Marie -- View this message in context: http://www.nabble.com/Error-in-Surv%28time%2C-status%29-%3A-Time-variable-is-not-numeric-tp21674025p21683673.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Sweave'ing Danish characters
Thank you, Duncan! It works perfectly!
Best regards,
Peter.
-Original Message-
From: Duncan Murdoch [mailto:murd...@stats.uwo.ca]
Sent: 27. januar 2009 13:04
To: Peter Jepsen
Cc: r-help@r-project.org
Subject: Re: [R] Sweave'ing Danish characters
On 26/01/2009 5:44 PM, Peter Jepsen wrote:
Hi,
I am writing an Sweave document and am using 'xtable' to make frequency
tables of diagnoses of people undergoing cholecystectomy. Some of these
diagnoses contain Danish characters (æ, ø, and å), and these characters
are all garbled in the Latex document after I run Sweave. The odd thing is,
everything looks absolutely right in the R console, and if I enter the same
Danish characters in a new variable, the new variable produces no problems?!
Therefore, I cannot offer a reproducible example, but I am hoping nonetheless
that someone can point me towards a solution.
This looks like an encoding problem: there are several different
standards for encoding non-ASCII characters. All of your tools have to
agree on the encoding.
To my eye it looks as though in the first case R is writing out UTF-8,
and whatever you are using to look at your .tex file is assuming latin1
(some Windows programs say ANSI, but I think that doesn't fully
specify the encoding: you also need a code page, which is set somewhere
in Windows control panel.)
The functions related to encodings in R are:
options(encoding=latin1) - set the default encoding
iconv(x, from=latin1, to=UTF-8) - re-encode entries, mapping each
character from one encoding to the other
Encoding(x) - display the encoding of each entry (unknown means ascii
or the native encoding for your platform)
Encoding(x) - latin1 - change the declared encoding, without
changing the bytes.
Duncan Murdoch
To illustrate:
library(xtable)
library(Hmisc)
rm(list=ls())
load(u:/kirurgi/cholecystit/Chol_oprenset.Rdata)
test2 - chol$nydiag[3] # This 3rd observation contains a diagnosis
with Danish characters (Kræft i fordøjelsessystemet, meaning
gastrointestinal cancer).
print(xtable(table(test2)))
% latex table generated in R 2.8.1 by xtable 1.5-4 package
% Mon Jan 26 23:31:37 2009
\begin{table}[ht]
\begin{center}
\begin{tabular}{rr}
\hline
test2 \\
\hline
Kræft i fordøjelsessystemet1 \\ # It looks right here, but in the
.tex-file it says Kræft i fordøjelsessystemet
\hline
\end{tabular}
\end{center}
\end{table}
print(xtable(table(Kræft i fordøjelsessystemet))) # This, on the other
hand, works like a charm.
% latex table generated in R 2.8.1 by xtable 1.5-4 package
% Mon Jan 26 23:36:53 2009
\begin{table}[ht]
\begin{center}
\begin{tabular}{rr}
\hline
V1 \\
\hline
Kræft i fordøjelsessystemet1 \\ # See, no problems here!
\hline
\end{tabular}
\end{center}
\end{table}
I am using Windows Vista 64-bit and MikTex 2.7.
Best regards,
Peter.
sessionInfo()
R version 2.8.1 (2008-12-22)
i386-pc-mingw32
locale:
LC_COLLATE=Danish_Denmark.1252;LC_CTYPE=Danish_Denmark.1252;LC_MONETARY=Danish_Denmark.1252;LC_NUMERIC=C;LC_TIME=Danish_Denmark.1252
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] Hmisc_3.4-4foreign_0.8-30 xtable_1.5-4
loaded via a namespace (and not attached):
[1] cluster_1.11.12 grid_2.8.1 lattice_0.17-20 tools_2.8.1
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] Sweave'ing Danish characters
On 26/01/2009 5:44 PM, Peter Jepsen wrote:
Hi,
I am writing an Sweave document and am using 'xtable' to make frequency tables of diagnoses of people
undergoing cholecystectomy. Some of these diagnoses contain Danish characters (æ, ø,
and å), and these characters are all garbled in the Latex document after I run Sweave. The odd
thing is, everything looks absolutely right in the R console, and if I enter the same Danish characters in a
new variable, the new variable produces no problems?! Therefore, I cannot offer a reproducible example, but I
am hoping nonetheless that someone can point me towards a solution.
This looks like an encoding problem: there are several different
standards for encoding non-ASCII characters. All of your tools have to
agree on the encoding.
To my eye it looks as though in the first case R is writing out UTF-8,
and whatever you are using to look at your .tex file is assuming latin1
(some Windows programs say ANSI, but I think that doesn't fully
specify the encoding: you also need a code page, which is set somewhere
in Windows control panel.)
The functions related to encodings in R are:
options(encoding=latin1) - set the default encoding
iconv(x, from=latin1, to=UTF-8) - re-encode entries, mapping each
character from one encoding to the other
Encoding(x) - display the encoding of each entry (unknown means ascii
or the native encoding for your platform)
Encoding(x) - latin1 - change the declared encoding, without
changing the bytes.
Duncan Murdoch
To illustrate:
library(xtable)
library(Hmisc)
rm(list=ls())
load(u:/kirurgi/cholecystit/Chol_oprenset.Rdata)
test2 - chol$nydiag[3] # This 3rd observation contains a diagnosis with Danish
characters (Kræft i fordøjelsessystemet, meaning gastrointestinal cancer).
print(xtable(table(test2)))
% latex table generated in R 2.8.1 by xtable 1.5-4 package
% Mon Jan 26 23:31:37 2009
\begin{table}[ht]
\begin{center}
\begin{tabular}{rr}
\hline
test2 \\
\hline
Kræft i fordøjelsessystemet1 \\# It looks right here, but in the .tex-file
it says Kræft i fordøjelsessystemet
\hline
\end{tabular}
\end{center}
\end{table}
print(xtable(table(Kræft i fordøjelsessystemet))) # This, on the other
hand, works like a charm.
% latex table generated in R 2.8.1 by xtable 1.5-4 package
% Mon Jan 26 23:36:53 2009
\begin{table}[ht]
\begin{center}
\begin{tabular}{rr}
\hline
V1 \\
\hline
Kræft i fordøjelsessystemet1 \\# See, no problems here!
\hline
\end{tabular}
\end{center}
\end{table}
I am using Windows Vista 64-bit and MikTex 2.7.
Best regards,
Peter.
sessionInfo()
R version 2.8.1 (2008-12-22)
i386-pc-mingw32
locale:
LC_COLLATE=Danish_Denmark.1252;LC_CTYPE=Danish_Denmark.1252;LC_MONETARY=Danish_Denmark.1252;LC_NUMERIC=C;LC_TIME=Danish_Denmark.1252
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] Hmisc_3.4-4foreign_0.8-30 xtable_1.5-4
loaded via a namespace (and not attached):
[1] cluster_1.11.12 grid_2.8.1 lattice_0.17-20 tools_2.8.1
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] Help
Since we have no idea what your data looks like, you can create a plot and then apply your own axis. You might also be able to rescale the x-axis data to be a percentage to start with. It is all doable depending on what you input looks like and what you want your output to look like. That is the beauty of R. On Mon, Jan 26, 2009 at 11:33 PM, Huihua Lu h...@mix.wvu.edu wrote: Hi, All, I want to construct a pareto chart for my data. My data has two columns: modules and defects(each module has several defects). Now I use pareto.chart to create a chart for defects which y axis is cumulative percentage of defect, and x axis is value of defect. That is not what I want. I need the x axis be percentage of modules. Is there any way that I can change the scale of x axis, that make x axis be percent scale for modules, like 0% to 100% modules. So that I can directly read the chart, is that 20% modules contains 80% defect. Or, is there any function that I can use to achive it directly? Thank you in advanced! Hedy __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Jim Holtman Cincinnati, OH +1 513 646 9390 What is the problem that you are trying to solve? __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] RCurl unable to download a particular web page -- what is so special about this web page?
Thank you Duncan. I remember seeing in your documentation that you have used this 'verbose=TRUE' argument in functions before when trying to see what is going on. This is good. However, I have not been able to get it to work for me. Does the output appear in R or do you use some other external window (i.e. MS DOS window?)? library(RCurl) my.url - 'http://www.nytimes.com/2009/01/07/technology/business-computing/07program.html?_r=2' getURL(my.url, verbose = TRUE) [1] I am having a problem with a new webpage (http://uk.youtube.com/) but if i can get this verbose to work, then i think i will be able to google the right action to take based on the information it gives. Many thanks for your time, C.C. On 26 Jan, 16:12, Duncan Temple Lang dun...@wald.ucdavis.edu wrote: clair.crossup...@googlemail.com wrote: Dear R-help, There seems to be a web page I am unable to download using RCurl. I don't understand why it won't download: library(RCurl) my.url - http://www.nytimes.com/2009/01/07/technology/business-computing/07pro...; getURL(my.url) [1] I like the irony that RCurl seems to have difficulties downloading an article about R. Good thing it is just a matter of additional arguments to getURL() or it would be bad news. The followlocation parameter defaults to FALSE, so getURL(my.url, followlocation = TRUE) gets what you want. The way I found this is getURL(my.url, verbose = TRUE) and take a look at the information being sent from R and received by R from the server. This gives * About to connect() towww.nytimes.comport 80 (#0) * Trying 199.239.136.200... * connected * Connected towww.nytimes.com(199.239.136.200) port 80 (#0) GET /2009/01/07/technology/business-computing/07program.html?_r=2 HTTP/1.1 Host:www.nytimes.com Accept: */* HTTP/1.1 301 Moved Permanently Server: Sun-ONE-Web-Server/6.1 Date: Mon, 26 Jan 2009 16:10:51 GMT Content-length: 0 Content-type: text/html Location:http://www.nytimes.com/glogin?URI=http://www.nytimes.com/2009/01/07/t... And the 301 is the critical thing here. D. Other web pages are ok to download but this is the first time I have been unable to download a web page using the very nice RCurl package. While i can download the webpage using the RDCOMClient, i would like to understand why it doesn't work as above please? library(RDCOMClient) my.url - http://www.nytimes.com/2009/01/07/technology/business-computing/07pro...; ie - COMCreate(InternetExplorer.Application) txt - list() ie$Navigate(my.url) NULL while(ie[[Busy]]) Sys.sleep(1) txt[[my.url]] - ie[[document]][[body]][[innerText]] txt $`http://www.nytimes.com/2009/01/07/technology/business-computing/ 07program.html?_r=2` [1] Skip to article Try Electronic Edition Log ... Many thanks for your time, C.C Windows Vista, running with administrator privileges. sessionInfo() R version 2.8.1 (2008-12-22) i386-pc-mingw32 locale: LC_COLLATE=English_United Kingdom.1252;LC_CTYPE=English_United Kingdom. 1252;LC_MONETARY=English_United Kingdom. 1252;LC_NUMERIC=C;LC_TIME=English_United Kingdom.1252 attached base packages: [1] stats graphics grDevices utils datasets methods base other attached packages: [1] RDCOMClient_0.92-0 RCurl_0.94-0 loaded via a namespace (and not attached): [1] tools_2.8.1 __ r-h...@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guidehttp://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ r-h...@r-project.org mailing listhttps://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guidehttp://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] RCurl unable to download a particular web page -- what is so special about this web page?
clair.crossup...@googlemail.com wrote: Thank you Duncan. I remember seeing in your documentation that you have used this 'verbose=TRUE' argument in functions before when trying to see what is going on. This is good. However, I have not been able to get it to work for me. Does the output appear in R or do you use some other external window (i.e. MS DOS window?)? The libcurl code typically defaults to print on the console. So on the Windows GUI, this will not show up. Using a shell (MS DOS window or Unix-like shell) should should cause the output to be displayed. A more general way however is to use the debugfunction option. d = debugGatherer() getURL(http://uk.youtube.com;, debugfunction = d$update, verbose = TRUE) When this completes, use d$value() and you have the entire contents that would be displayed on the console. D. library(RCurl) my.url - 'http://www.nytimes.com/2009/01/07/technology/business-computing/07program.html?_r=2' getURL(my.url, verbose = TRUE) [1] I am having a problem with a new webpage (http://uk.youtube.com/) but if i can get this verbose to work, then i think i will be able to google the right action to take based on the information it gives. Many thanks for your time, C.C. On 26 Jan, 16:12, Duncan Temple Lang dun...@wald.ucdavis.edu wrote: clair.crossup...@googlemail.com wrote: Dear R-help, There seems to be a web page I am unable to download using RCurl. I don't understand why it won't download: library(RCurl) my.url - http://www.nytimes.com/2009/01/07/technology/business-computing/07pro...; getURL(my.url) [1] I like the irony that RCurl seems to have difficulties downloading an article about R. Good thing it is just a matter of additional arguments to getURL() or it would be bad news. The followlocation parameter defaults to FALSE, so getURL(my.url, followlocation = TRUE) gets what you want. The way I found this is getURL(my.url, verbose = TRUE) and take a look at the information being sent from R and received by R from the server. This gives * About to connect() towww.nytimes.comport 80 (#0) * Trying 199.239.136.200... * connected * Connected towww.nytimes.com(199.239.136.200) port 80 (#0) GET /2009/01/07/technology/business-computing/07program.html?_r=2 HTTP/1.1 Host:www.nytimes.com Accept: */* HTTP/1.1 301 Moved Permanently Server: Sun-ONE-Web-Server/6.1 Date: Mon, 26 Jan 2009 16:10:51 GMT Content-length: 0 Content-type: text/html Location:http://www.nytimes.com/glogin?URI=http://www.nytimes.com/2009/01/07/t... And the 301 is the critical thing here. D. Other web pages are ok to download but this is the first time I have been unable to download a web page using the very nice RCurl package. While i can download the webpage using the RDCOMClient, i would like to understand why it doesn't work as above please? library(RDCOMClient) my.url - http://www.nytimes.com/2009/01/07/technology/business-computing/07pro...; ie - COMCreate(InternetExplorer.Application) txt - list() ie$Navigate(my.url) NULL while(ie[[Busy]]) Sys.sleep(1) txt[[my.url]] - ie[[document]][[body]][[innerText]] txt $`http://www.nytimes.com/2009/01/07/technology/business-computing/ 07program.html?_r=2` [1] Skip to article Try Electronic Edition Log ... Many thanks for your time, C.C Windows Vista, running with administrator privileges. sessionInfo() R version 2.8.1 (2008-12-22) i386-pc-mingw32 locale: LC_COLLATE=English_United Kingdom.1252;LC_CTYPE=English_United Kingdom. 1252;LC_MONETARY=English_United Kingdom. 1252;LC_NUMERIC=C;LC_TIME=English_United Kingdom.1252 attached base packages: [1] stats graphics grDevices utils datasets methods base other attached packages: [1] RDCOMClient_0.92-0 RCurl_0.94-0 loaded via a namespace (and not attached): [1] tools_2.8.1 __ r-h...@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guidehttp://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ r-h...@r-project.org mailing listhttps://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guidehttp://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] retrieving variables values from a function
Hello experts! Is there a way to send an internal variable from a function to the workspace, besides the function output, of course Thanks!! D. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] glm StepAIC with all interactions and update to remove a term vs. glm specifying all but a few terms and stepAIC
At 12:49 26/01/2009, Robert Michael Inman wrote:
Problem:
I am sorting through model selection process for first time and want to make
sure that I have used glm, stepAIC, and update correctly. Something is
strange because I get a different result between:
1) a glm of 12 predictor variables followed by a stepAIC where all
interactions are considered and then an update to remove one specific
interaction.
vs.
2) entering all the terms individually in a glm (exept the one that I
removed with update and 4 others like it but which did not make it to final
model anyway), and then running stepAIC.
I am not the world's leading expert on this area but nobody else
seems to have replied yet so here goes:
1 - stepwise methods capitalise on chance features of your dataset
and so starting from a slightly different place may give different
results. For instance if you do all possible subsets then the 'best'
subset of size k is not guaranteed to include the members of the
'best' subset of size j (jk) and indeed may not include any of them.
2 - the lack of significance of some predictors in the last model is
probably because stepAIC respects marginality, certainly MASS implies this.
You might find
@book{miller90,
author = Miller, A J,
title = Subset selection in regression,
year = 1990,
publisher = Chapman and Hall,
address = London,
keywords= {stepwise}
}
helpful. And MASS of course since the package is support for the book.
Question:
Why do these processes not yield same model?
Here are all the details if helpful:
I start with 12 potential predictor variables, 7 primary terms and 5
additional that are I(primary_terms^2). I run a glm for these 12 and then
do stepAIC (BIC actually) both directions. The scope argument is
scope=list(upper=~.^2,lower=NULL). This means there are 78 predictor terms
considered, the 12 primary terms and 66 interactions [n(n+1)/2]. I see this
with trace=T also. Here is the code used:
glm1-glm(formula = PRESENCE == 1 ~ SNOW + I(SNOW^2) + POP_DEN + ROAD_DE
+ ADJELEV + I(ADJELEV^2) + TRI + I(TRI^2) + EDGE + I(EDGE^2) + TREECOV +
I(TREECOV^2),family = binomial, data = wolv)
summary(glm1)
library(MASS)
stepglm2-stepAIC(glm1,scope=list(upper=~.^2,lower=NULL),
trace=T,k=log(4828),direction=both)
summary(stepglm2)
extractAIC(stepglm2,k=log(4828))
This results in a 15 term model with a BIC of 3758.659
Coefficients:
Estimate Std. Error z value Pr(|z|)
(Intercept) -4.983e+01 9.263e+00 -5.379 7.50e-08 ***
SNOW 6.085e-02 8.641e-03 7.041 1.90e-12 ***
ROAD_DE -5.637e-01 1.192e-01 -4.730 2.24e-06 ***
ADJELEV2.880e-02 7.457e-03 3.863 0.000112 ***
I(ADJELEV^2) -4.038e-06 1.487e-06 -2.715 0.006618 **
TRI5.675e-02 1.081e-02 5.248 1.54e-07 ***
I(TRI^2) -1.713e-03 4.243e-04 -4.036 5.43e-05 ***
EDGE 6.418e-03 1.697e-03 3.782 0.000156 ***
TREECOV1.680e-01 2.929e-02 5.735 9.76e-09 ***
SNOW:ADJELEV -4.313e-05 6.935e-06 -6.219 5.00e-10 ***
ADJELEV:TREECOV -6.628e-05 1.161e-05 -5.711 1.13e-08 ***
SNOW:I(ADJELEV^2) 7.437e-09 1.384e-09 5.373 7.74e-08 ***
TRI:I(TRI^2) 1.321e-06 3.419e-07 3.863 0.000112 ***
I(ADJELEV^2):I(TRI^2) -2.127e-10 5.745e-11 -3.702 0.000214 ***
ADJELEV:I(TRI^2) 1.029e-06 3.004e-07 3.424 0.000617 ***
SNOW:TRI 1.057e-05 3.372e-06 3.135 0.001721 **
The final model included a the TRI:I(TRI^2) term, which is effectively a
cubic function. So this was removed because cubic's were not considered for
all variables. I used update to remove TRI:I(TRI^2). Code:
stepglm3-update(stepglm2,~.-TRI:I(TRI^2),trace=T)
summary(stepglm3)
extractAIC(stepglm3,k=log(4828))
This results in a 14 term model with a BIC of 3770.172. The BIC is a little
higher, but the cubic term improved fit and is no longer in, so expected.
Coefficients:
Estimate Std. Error z value Pr(|z|)
(Intercept) -5.329e+01 9.267e+00 -5.750 8.92e-09 ***
SNOW 6.241e-02 8.695e-03 7.178 7.06e-13 ***
ROAD_DE -5.756e-01 1.184e-01 -4.863 1.16e-06 ***
ADJELEV3.233e-02 7.452e-03 4.338 1.44e-05 ***
I(ADJELEV^2) -4.724e-06 1.487e-06 -3.177 0.001489 **
TRI1.834e-02 5.402e-03 3.395 0.000687 ***
I(TRI^2) -1.122e-03 3.920e-04 -2.863 0.004190 **
EDGE 6.344e-03 1.690e-03 3.754 0.000174 ***
TREECOV1.745e-01 2.923e-02 5.969 2.39e-09 ***
SNOW:ADJELEV -4.444e-05 6.984e-06 -6.363 1.98e-10 ***
ADJELEV:TREECOV -6.885e-05 1.160e-05 -5.937 2.90e-09 ***
SNOW:I(ADJELEV^2) 7.681e-09 1.395e-09 5.506 3.67e-08 ***
I(ADJELEV^2):I(TRI^2)
Re: [R] retrieving variables values from a function
I guess by workspace you mean global environment. I believe this is generally considered a bad practice, but see ?assign and ?- baptiste On 27 Jan 2009, at 13:54, diego Diego wrote: Hello experts! Is there a way to send an internal variable from a function to the workspace, besides the function output, of course Thanks!! D. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. _ Baptiste Auguié School of Physics University of Exeter Stocker Road, Exeter, Devon, EX4 4QL, UK Phone: +44 1392 264187 http://newton.ex.ac.uk/research/emag __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Windows, gsview: corrupted bbox of .e?ps graphics
System: [Windows XP Pro, R 2.8.0 and R 2.81.]
When I do R graphics on Windows and want .eps output for LaTeX, with *tight
bounding boxes* I usually do as follows from Rgui:
- Save as Postscript ... myfig.ps
- Open myfig.ps in Gsview, select PS to EPS, Save as ... myfig.eps
This always worked before, and gave me .eps files that did not include much
more white space than I like, didn't require fiddling with par(), nor using
\includegraphics[trim=]{myfig}.
But now I've encountered a problem that prevents this workflow from
working --- when I now open *any* R-generated-this-way .ps or .eps file,
with GSview (v 4.9), the bounding box of the graphic is corrupted--
i.e., the bounding box I see cuts off the top half of the figure.
See:
http://euclid.psych.yorku.ca/SCS/Private/agree-MS-buggged.eps
for
the example produced below.
This does view OK with gv on linux, however, the usual tool I use on
linux to find new, tight bbox won't work on this file:
%psfixbb -l agree-MS-bugged.eps
psfixbb: Probloem with new %%BoundingBox: -4 -4 616 796
What caused this was an example from vcd, that I wanted in .eps form.
What I got from
my usual method of Save as Postscript was not at all pleasing--- font
sizes in the
.ps image were way larger than what appeared on the screen. So, I added a
postscript() command to the example in ?agreementplot
library(vcd)
data(MSPatients)
## best visualized using a resized device, e.g. using:
#get(getOption(device))(width = 12)
dev.new()
postscript(file=agree-MS.ps, paper=special, horizontal=FALSE,
width=6.0, height=3.0)
pushViewport(viewport(layout = grid.layout(ncol = 2)))
pushViewport(viewport(layout.pos.col = 2))
agreementplot(t(MSPatients[,,1]), main = Winnipeg Patients,
newpage = FALSE)
popViewport()
pushViewport(viewport(layout.pos.col = 1))
agreementplot(t(MSPatients[,,2]), main = New Orleans Patients,
newpage = FALSE)
popViewport(2)
dev.off()
Ever since, when I use gsview on Win to view *any* .eps file
generated via Save as Postscript from the R graphics window,
I get similar behavior. This happens in a fresh R session, as well as
with R 2.8.1. It doesn't happen with Sweave-generated
graphics (but I haven't tried the above example in Sweave).
Any ideas on what is wrong and how I can fix this?
-Michael
--
Michael Friendly Email: friendly AT yorku DOT ca
Professor, Psychology Dept.
York University Voice: 416 736-5115 x66249 Fax: 416 736-5814
4700 Keele Streethttp://www.math.yorku.ca/SCS/friendly.html
Toronto, ONT M3J 1P3 C
NADA
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
[R] plot Ticks
Is there a way to force the number of ticks along an axis ? I read the on-line documentation and tried many combinations of all available parameters from functions par(), axTicks(), axis(), plot() ... but no luck ! Thank you very much, Maura tutti i telefonini TIM! [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] plot Ticks
Will something like this do what you want? x - 1:10 y=rnorm(10) plot(x,y,xaxt=n) axis(1,at= seq(.5,10,.5)) --- On Tue, 1/27/09, mau...@alice.it mau...@alice.it wrote: From: mau...@alice.it mau...@alice.it Subject: [R] plot Ticks To: r-h...@stat.math.ethz.ch Received: Tuesday, January 27, 2009, 9:13 AM Is there a way to force the number of ticks along an axis ? I read the on-line documentation and tried many combinations of all available parameters from functions par(), axTicks(), axis(), plot() ... but no luck ! Thank you very much, Maura tutti i telefonini TIM! [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ [[elided Yahoo spam]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] plot Ticks
Try this: plot(1:20) axis(3, at=seq(0,20), label=FALSE) A better description of your plot would be useful if ?axis is not enough to help you out. hope this helps, baptiste On 27 Jan 2009, at 14:13, mau...@alice.it wrote: Is there a way to force the number of ticks along an axis ? I read the on-line documentation and tried many combinations of all available parameters from functions par(), axTicks(), axis(), plot() ... but no luck ! Thank you very much, Maura tutti i telefonini TIM! [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. _ Baptiste Auguié School of Physics University of Exeter Stocker Road, Exeter, Devon, EX4 4QL, UK Phone: +44 1392 264187 http://newton.ex.ac.uk/research/emag __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Error in Surv(time, status)
I used following command: frailtyPenal(Surv(time, status) ~var1 + cluster(family), Frailty=TRUE ,n.knots=8, kappa1=1500, + cross.validation=FALSE) And got this error : Error in Surv(time, status) : Time variable is not numeric In addition: Warning message: In is.na(time) : is.na() applied to non-(list or vector) of type 'closure' --- It looks like you forgot data= in your model statement, which makes 'time' refer to the function of that name. Terry T. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] plot Ticks
you could set them manually, if thats what you're looking for plot(1:10, axes = F) axis(1, at = seq(1,10 , length = 3)) mau...@alice.it schrieb: Is there a way to force the number of ticks along an axis ? I read the on-line documentation and tried many combinations of all available parameters from functions par(), axTicks(), axis(), plot() ... but no luck ! Thank you very much, Maura tutti i telefonini TIM! [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] retrieving variables values from a function
Just have it as a return value and then assign it when you return. You should have have side effects coming from your functions. There are ways of doing it, but it you think you have to, think about it again. On Tue, Jan 27, 2009 at 8:54 AM, diego Diego dhab...@gmail.com wrote: Hello experts! Is there a way to send an internal variable from a function to the workspace, besides the function output, of course Thanks!! D. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Jim Holtman Cincinnati, OH +1 513 646 9390 What is the problem that you are trying to solve? __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] retrieving variables values from a function
Assuming you _really_ want to do this check out '-' and closures... And remember that your code will probably make other people cry when they go to debug it. Krzysztof Sent via BlackBerry by ATT -Original Message- From: jim holtman jholt...@gmail.com Date: Tue, 27 Jan 2009 09:43:17 To: diego Diegodhab...@gmail.com Cc: r-help@r-project.org Subject: Re: [R] retrieving variables values from a function Just have it as a return value and then assign it when you return. You should have have side effects coming from your functions. There are ways of doing it, but it you think you have to, think about it again. On Tue, Jan 27, 2009 at 8:54 AM, diego Diego dhab...@gmail.com wrote: Hello experts! Is there a way to send an internal variable from a function to the workspace, besides the function output, of course Thanks!! D. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Jim Holtman Cincinnati, OH +1 513 646 9390 What is the problem that you are trying to solve? __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Problem with RMA using limma, oligo and pdInfoBuilder packages
Hi, I am a Ph.D. student from Québec, Canada. Im a beginner with R and Bioconductor. Until now the only experience I have is in analyzing microarray data using affy and limma packages. Now I am trying to analyze Rat Gene 10 st arrays and I would like to run RMA analysis and Smyth moderated t test on those arrays. Since no cdf official package is available for those arrays, after reading many of the questions and responses on this mailing list, I decided to use pdInfoBuilder, oligo and limma packages to run analysis. The problem is, at the end, I get expression and differential expression measured for all probe separately but not the calculated expression representing all probe of each gene. When I run RMA, I got only two steps, Background correcting and Normalizing but not Calculating expression. Do you know how I can get differential expression calculated for each gene? I dont know if the problem is in the package I built or if I can use some code to answer this question. I list all codes used to build and install the package pd.ragene.1.0.st.v1 and used to analyze expression arrays below. Many thanks for your help, Anne-Marie Madore ## building the package library(Biobase) Loading required package: tools Welcome to Bioconductor Vignettes contain introductory material. To view, type 'openVignette()'. To cite Bioconductor, see 'citation(Biobase)' and for packages 'citation(pkgname)'. library(pdInfoBuilder) Loading required package: RSQLite Loading required package: DBI Loading required package: affxparser Loading required package: oligo Loading required package: splines Loading required package: preprocessCore Loading required package: AnnotationDbi Loading required package: oligoClasses oligo Package - Series 1.5.x setwd(D:/Anne-Marie/Doctorat/puces ADN macrophages/puces rat/Annie Dube/Analyse) transFile - RaGene-1_0-st-v1.na27.rn4.transcript.csv1/RaGene-1_0-st-v1.na27.rn4.transcr ipt.csv probeFile - RaGene-1_0-st-v1.probe.tab/RaGene-1_0-st-v1.probe.tab clfFile - RaGene-1_0-st-v1.r4.clf/RaGene-1_0-st-v1.r4.clf pgfFile - RaGene-1_0-st-v1.r4.pgf/RaGene-1_0-st-v1.r4.pgf pkg - new(AffyGenePDInfoPkgSeed, author=Anne-Marie Madore, email=anne-marie.mador...@ulaval.ca, version=0.0.1, + genomebuild=RefSeq April 3, 2007, GenBank® January 25, 2007, Rat Ensembl transcripts April 3, 2007 , + biocViews=AnnotationData, pgfFile=pgfFile, clfFile=clfFile, transFile=transFile, probeFile=probeFile) makePdInfoPackage(pkg, destDir=.) Creating package in ./pd.ragene.1.0.st.v1 loadUnitsByBatch took 50.51 sec loadAffyCsv took 12.73 sec loadAffySeqCsv took 57.62 sec DB sort, index creation took 24.75 sec [1] TRUE Warning messages: 1: In is.na(x) : is.na() applied to non-(list or vector) of type 'NULL' 2: In is.na(x) : is.na() applied to non-(list or vector) of type 'NULL' ## installing the package in cmd command shell Microsoft Windows [version 6.0.6001] Copyright (c) 2006 Microsoft Corporation. Tous droits réservés. C:\Users\Anne-Marie Madorecd c:\Program Files\R\R-2.8.1\bin c:\Program Files\R\R-2.8.1\binR CMD INSTALL pd.ragene.1.0.st.v1 installing to 'c:/PROGRA~1/R/R-28~1.1/library' -- Making package pd.ragene.1.0.st.v1 adding build stamp to DESCRIPTION installing NAMESPACE file and metadata installing R files installing inst files preparing package pd.ragene.1.0.st.v1 for lazy loading Loading required package: RSQLite Loading required package: DBI Loading required package: oligoClasses Loading required package: Biobase Loading required package: tools Welcome to Bioconductor Vignettes contain introductory material. To view, type 'openVignette()'. To cite Bioconductor, see 'citation(Biobase)' and for packages 'citation(pkgname)'. no man files in this package installing indices installing help adding MD5 sums * DONE (pd.ragene.1.0.st.v1) ## If I run a check (R CMD check pd.ragene.st.v1) I get three warning messages and one note: 1. * checking R files for non-ASCII characters ... WARNING Found the following files with non-ASCII characters: all.R Portable packages must use only ASCII characters in their R code, except perhaps in comments. 2. * checking whether the name space can be loaded with stated dependencies ... WARNING Error in initDbConnection() : could not find function dbConnect Error: .onLoad failed in 'loadNamespace' for 'pd.ragene.1.0.st.v1' Execution halted A namespace must be able to be loaded with just the base namespace loaded: otherwise if the namespace gets loaded by a saved object, the session will be unable to start. Probably some imports need to be declared in the NAMESPACE file. 3. * checking R code for possible problems ... NOTE closeDb: no visible binding for global variable 'dbCon' 4. * checking for missing documentation entries ... WARNING Undocumented code objects:
[R] Mixed Effects Model with Different Group Sizes
Hi R user, I can not find the answer for a simple problem. I want to test the main effects and interactions of a 2 by 2 factorial design (Experiment I) and 2 by 2 by 2 factorial design (Experiment II) with unequal group sizes and one or two within-subject factors. 2 by 2 Design (Experiment I): Factor A is a fixed grouping factor with different group sizes for group A1 and A2 (reflecting the different appearance of the factor levels within the population). Factor B is a fixed “within-subject” factor and has two levels: The means of conditions B1 and B2 for each subject. Thereby condition B1 has been measured more frequently than condition B2. Each subject in group A1 and A2 was measured under the conditions B1 and B2. Experiment II has an additional “within-subject” factor C. Thereby B1 and B2 were measured under conditions C1 and C2. From my current readings I suggest for Experiment I: library(lme4) model1 - lmer (beta ~ B - 1+(1| subject), data=data.ex,REML=FALSE) model2 - lmer (beta ~ A - 1 +(1| subject), data=data.ex,REML=FALSE) model3 - lmer (beta ~ A + B - 1 +(1| subject),data=data.ex,REML=FALSE) model4 - lmer (beta ~ A * B - 1 +(1| subject), data=data.ex,REML=FALSE) anova(model1,model3) #Main effect factor A anova(model2,model3) #Main effect factor B anova(model3,model4) #Interaction factor A x B Is this correct (error strata for unequal group sizes)? Do you suggest different solutions? Does the solution for experiment II follow the same principle of error strata? Many thanks, Carsten __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Goodness of fit for gamma distributions
Thanks for that Remko, but im slightly confused because isnt this testing the goodness of fit of 2 slightly different gamma distributions, not of how well a gamma distribution is representing the data. e.g. data.vec-as.vector(data) (do some mle to find the parameters of a gamma distribution for data.vec) xrarea-seq(-2,9,0.05) yrarea-dgamma(xrarea,shape=7.9862,rate=2.6621) so now yrarea is the gamma distribution and i want to compare it with data.vec to see how well it fits. regards, Dann Remko Duursma-2 wrote: Hi Dann, there is probably a better way to do this, but this works anyway: # your data gamdat - rgamma(1, shape=1, rate=0.5) # comparison to gamma: gamsam - rgamma(1, shape=1, rate=0.6) qqplot(gamsam,gamdat) abline(0,1) greetings Remko - Remko Duursma Post-Doctoral Fellow Centre for Plant and Food Science University of Western Sydney Hawkesbury Campus Richmond NSW 2753 Dept of Biological Science Macquarie University North Ryde NSW 2109 Australia Mobile: +61 (0)422 096908 On Tue, Jan 27, 2009 at 3:38 AM, Dan31415 d.m.mitch...@reading.ac.uk wrote: I'm looking for goodness of fit tests for gamma distributions with large data sizes. I have a matrix with around 10,000 data values in it and i have fitted a gamma distribution over a histogram of the data. The problem is testing how well that distribution fits. Chi-squared seems to be used more for discrete distributions and kolmogorov-smirnov seems that large sample sizes make it had to evaluate the D statistic. Also i haven't found a qq plot for gamma, although i think this might be an appropriate test. in summary -is there a gamma goodness of fit test that doesnt depend on the sample size? -is there a way of using qqplot for gamma distributions, if so how would you calculate it from a matrix of data values? regards, Dann -- View this message in context: http://www.nabble.com/Goodness-of-fit-for-gamma-distributions-tp21668711p21668711.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- View this message in context: http://www.nabble.com/Goodness-of-fit-for-gamma-distributions-tp21668711p21686095.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Error in Surv(time, status) : Time variable is not numeric
Thank you for your reply. Even when I put data=.. in my model statement, I get the same error. Chris Andrews wrote: R is finding the R function 'time' rather than your variable time in your dataset. Perhaps adding 'data=mydata' to your function call will resolve your issue. Chris Braem M wrote: Dear, I want to analyze two-level survival data using a shared frailty model, for which I want to use the R package 'Frailtypack, proposed by Rondeau et al. The dataset was built using SAS software. I also tried to change the format using SPSS and Excell. My (reduced) dataset has following column names: ID entrytimestatusfamily var1 I used following command: frailtyPenal(Surv(time, status) ~var1 + cluster(family), Frailty=TRUE ,n.knots=8, kappa1=1500, + cross.validation=FALSE) And got this error : Error in Surv(time, status) : Time variable is not numeric In addition: Warning message: In is.na(time) : is.na() applied to non-(list or vector) of type 'closure' I think R transforms the data when importing into R, so that the observations are not numeric anymore. Does anyone know how to handle this problem? Thanks, Marie -- View this message in context: http://www.nabble.com/Error-in-Surv%28time%2C-status%29-%3A-Time-variable-is-not-numeric-tp21674025p21686806.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Creating list or numeric vectors out of selected columns of row oriented data
I am just assuming this can be done, but I have not gotten close to making it happen. I have a data file with about 1 million rows with 1470 unique subjects. Each row represents a small set of observations made on a specific date for a single subject. I would like to transform the data so that I have an R object with a single entry for each subject and start date and vectors for the observation dates and the observations. The data are something like the following where for each subject the subject_id does not change and the start_date does not change, but the obeservation_date and the three different observations change between rows. (There is one row for each day for each subject over a three year period although some entered the study late): 'subject_id', 'start_date','observation_date','weight_obs', 'activity_obs','calories_obs' 1,'1/1/2005','1/1/2005',3.26,'a',93 1,'1/1/2005','1/2/2005',3.22,'o',85 1,'1/1/2005','1/3/2005',3.28,'o',91 ... 1,'1/1/2005','12/31/2008',4.38,'h',102 2,'2/13/2005','2/13/2005',3.02,'l',80 2,'2/13/2005','2/14/2005',3.08,'j',85 ... Any guidance is appreciated. R. Mark Sharp, Ph.D. Director of Primate Records Database Southwest National Primate Center Southwest Foundation for Biomedical Research P.O. Box 760549 San Antonio, TX 78245-0549 Telephone: (210)258-9476 e-mail: msh...@sfbr.org [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] optim() and ARIMA
dhabby wrote: Last week I run in to a lot a problems triyng to fit an ARIMA model to a time series. The problem is that the internal process of the arima function call function optim to estimate the model parameters, so far so good... but my data presents a problem with the default method BFGS of the optim function, the output error looks like this: Error en optim(init[mask], armafn, method = BFGS, hessian = TRUE, control = optim.control, : non-finite finite-difference value [7] I don't know much about the calls from ARIMA to optim, but when I modified Fletcher's 1970 VM method (called BFGS in R), I was aiming to make it extremely robust to messy functions. (I had machinery with only 6 hex digits for floating point and no guard digits. Furthermore, only 8K bytes for program AND data. This was 1975/6.) However, BFGS needs gradients, and it is clear there's some sort of trouble approximating them with finite differences. Changing optimization methods might succeed on this occasion, but your problem is more likely bad scaling or some setup where your ARIMA model essentially is not sensible. Can you duplicate the message on a small example problem? John Nash __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] RCurl unable to download a particular web page -- what is so special about this web page?
Thank you. The output i get from that example is below:
d = debugGatherer()
getURL(http://uk.youtube.com;,
+ debugfunction = d$update, verbose = TRUE )
[1]
d$value()
text
About to connect() to uk.youtube.com port 80 (#0)\n Trying
208.117.236.72... connected\nConnected to uk.youtube.com
(208.117.236.72) port 80 (#0)\nConnection #0 to host uk.youtube.com
left intact\n
headerIn
HTTP/1.1 400 Bad Request\r\nVia: 1.1 PFO-FIREWALL\r\nConnection: Keep-
Alive\r\nProxy-Connection: Keep-Alive\r\nTransfer-Encoding: chunked\r
\nExpires: Tue, 27 Apr 1971 19:44:06 EST\r\nDate: Tue, 27 Jan 2009
15:31:25 GMT\r\nContent-Type: text/plain\r\nServer: Apache\r\nX-
Content-Type-Options: nosniff\r\nCache-Control: no-cache\r
\nCneonction: close\r\n\r\n
headerOut
GET / HTTP/1.1\r\nHost: uk.youtube.com\r\nAccept: */*\r\n\r\n
dataIn
0\r\n\r\n
dataOut
So the critical information from this is the '400 Bad Request'. A
Google search defines this for me as:
The request could not be understood by the server due to malformed
syntax. The client SHOULD NOT repeat the request without
modifications.
looking through sort(both listCurlOptions() and
http://curl.haxx.se/libcurl/c/curl_easy_setopt.htm) doesn't really
help me this time (unless i missed something). Any advice?
Thank you for your time,
C.C
P.S. I can get the d/l to work if i use:
toString(readLines(http://www.uk.youtube.com;))
[1] html, \thead, \t\ttitleOpenDNS/title, \t/head, ,
\tbody id=\mainbody\ onLoad=\testforbanner();\ style=\margin:
0px;\, \t\tscript language=\JavaScript\, \t\t\tfunction
testforbanner() {, \t\t\t\tvar width;, \t\t\t\tvar height;, \t\t\t
\tvar x = 0;, \t\t\t\tvar isbanner = false;, \t\t\t\tvar bannersizes =
new Array(16), \t\t\t\tbannersizes[0] = [etc]
On 27 Jan, 13:52, Duncan Temple Lang dun...@wald.ucdavis.edu wrote:
clair.crossup...@googlemail.com wrote:
Thank you Duncan.
I remember seeing in your documentation that you have used this
'verbose=TRUE' argument in functions before when trying to see what is
going on. This is good. However, I have not been able to get it to
work for me. Does the output appear in R or do you use some other
external window (i.e. MS DOS window?)?
The libcurl code typically defaults to print on the console.
So on the Windows GUI, this will not show up. Using
a shell (MS DOS window or Unix-like shell) should
should cause the output to be displayed.
A more general way however is to use the debugfunction
option.
d = debugGatherer()
getURL(http://uk.youtube.com;,
debugfunction = d$update, verbose = TRUE)
When this completes, use
d$value()
and you have the entire contents that would be displayed on the console.
D.
library(RCurl)
my.url -
'http://www.nytimes.com/2009/01/07/technology/business-computing/07pro...
getURL(my.url, verbose = TRUE)
[1]
I am having a problem with a new webpage (http://uk.youtube.com/) but
if i can get this verbose to work, then i think i will be able to
google the right action to take based on the information it gives.
Many thanks for your time,
C.C.
On 26 Jan, 16:12, Duncan Temple Lang dun...@wald.ucdavis.edu wrote:
clair.crossup...@googlemail.com wrote:
Dear R-help,
There seems to be a web page I am unable to download using RCurl. I
don't understand why it won't download:
library(RCurl)
my.url -
http://www.nytimes.com/2009/01/07/technology/business-computing/07pro...;
getURL(my.url)
[1]
I like the irony that RCurl seems to have difficulties downloading an
article about R. Good thing it is just a matter of additional arguments
to getURL() or it would be bad news.
The followlocation parameter defaults to FALSE, so
getURL(my.url, followlocation = TRUE)
gets what you want.
The way I found this is
getURL(my.url, verbose = TRUE)
and take a look at the information being sent from R
and received by R from the server.
This gives
* About to connect() towww.nytimes.comport80 (#0)
* Trying 199.239.136.200... * connected
* Connected towww.nytimes.com(199.239.136.200) port 80 (#0)
GET /2009/01/07/technology/business-computing/07program.html?_r=2
HTTP/1.1
Host:www.nytimes.com
Accept: */*
HTTP/1.1 301 Moved Permanently
Server: Sun-ONE-Web-Server/6.1
Date: Mon, 26 Jan 2009 16:10:51 GMT
Content-length: 0
Content-type: text/html
Location:http://www.nytimes.com/glogin?URI=http://www.nytimes.com/2009/01/07/t...
And the 301 is the critical thing here.
D.
Other web pages are ok to download but this is the first time I have
been unable to download a web page using the very nice RCurl package.
While i can download the webpage using the RDCOMClient, i would like
to understand why it doesn't work as above please?
library(RDCOMClient)
my.url -
http://www.nytimes.com/2009/01/07/technology/business-computing/07pro...;
ie - COMCreate(InternetExplorer.Application)
txt - list()
Re: [R] optim() and ARIMA
Surely, this sounds like a bug in the optim function. The rule of thumb with ts data is to scale so that data have mean 0 and unit variance and then fit a) for non-seasonal data the IMA model (0,1,1); and b) for seasonal data so-called Airline Model (0,1,1)X(0,1,1)S see for example A course in TS analysis by Daniel Pena. These can usually be fit to most series unless they are very irregular (e.g. mainly zeros) and then Box-Jenkins ARIMA is inappropriate. The fit of any other model is then judged against these two models. If a proposed model fit fails it is usual report the fault and then to step over that model and try the next in sequence - optim should do this, if not then it's a code fault. Very often with nasty series only the 2 basic models can be fit and these results are reported - but for these the parameters may not even fully converge. The moral of all this is that automatic modelling methods are handy but not a substitute for looking at the acf and choosing an appropriate parameterisation. Can someone say whether optim has automatic outlier detection - this is a must have with this type of procedure. Gerard nas...@uottawa.ca Sent by: r-help-boun...@r- To project.org r-help@r-project.org cc 27/01/2009 15:45 Subject Re: [R] optim() and ARIMA dhabby wrote: Last week I run in to a lot a problems triyng to fit an ARIMA model to a time series. The problem is that the internal process of the arima function call function optim to estimate the model parameters, so far so good... but my data presents a problem with the default method BFGS of the optim function, the output error looks like this: Error en optim(init[mask], armafn, method = BFGS, hessian = TRUE, control = optim.control, : non-finite finite-difference value [7] I don't know much about the calls from ARIMA to optim, but when I modified Fletcher's 1970 VM method (called BFGS in R), I was aiming to make it extremely robust to messy functions. (I had machinery with only 6 hex digits for floating point and no guard digits. Furthermore, only 8K bytes for program AND data. This was 1975/6.) However, BFGS needs gradients, and it is clear there's some sort of trouble approximating them with finite differences. Changing optimization methods might succeed on this occasion, but your problem is more likely bad scaling or some setup where your ARIMA model essentially is not sensible. Can you duplicate the message on a small example problem? John Nash __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. ** The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. It is the policy of the Department of Justice, Equality and Law Reform and the Agencies and Offices using its IT services to disallow the sending of offensive material. Should you consider that the material contained in this message is offensive you should contact the sender immediately and also mailminder[at]justice.ie. Is le haghaidh an duine nó an eintitis ar a bhfuil sí dírithe, agus le haghaidh an duine nó an eintitis sin amháin, a bheartaítear an fhaisnéis a tarchuireadh agus féadfaidh sé go bhfuil ábhar faoi rún agus/nó faoi phribhléid inti. Toirmisctear aon athbhreithniú, atarchur nó leathadh a dhéanamh ar an bhfaisnéis seo, aon úsáid eile a bhaint aisti nó aon ghníomh a
Re: [R] Creating list or numeric vectors out of selected columns of row oriented data
Yes the data can probably be easily transformed, but you would have to provide an example of what the input looks like to understand what has to be done with the data and how variable it might be so we can understand how we might have to parse the data from the input. Are the missing days supposed to be filled with with NAs? On Tue, Jan 27, 2009 at 10:36 AM, R. Mark Sharp msh...@sfbr.org wrote: I am just assuming this can be done, but I have not gotten close to making it happen. I have a data file with about 1 million rows with 1470 unique subjects. Each row represents a small set of observations made on a specific date for a single subject. I would like to transform the data so that I have an R object with a single entry for each subject and start date and vectors for the observation dates and the observations. The data are something like the following where for each subject the subject_id does not change and the start_date does not change, but the obeservation_date and the three different observations change between rows. (There is one row for each day for each subject over a three year period although some entered the study late): 'subject_id', 'start_date','observation_date','weight_obs', 'activity_obs','calories_obs' 1,'1/1/2005','1/1/2005',3.26,'a',93 1,'1/1/2005','1/2/2005',3.22,'o',85 1,'1/1/2005','1/3/2005',3.28,'o',91 ... 1,'1/1/2005','12/31/2008',4.38,'h',102 2,'2/13/2005','2/13/2005',3.02,'l',80 2,'2/13/2005','2/14/2005',3.08,'j',85 ... Any guidance is appreciated. R. Mark Sharp, Ph.D. Director of Primate Records Database Southwest National Primate Center Southwest Foundation for Biomedical Research P.O. Box 760549 San Antonio, TX 78245-0549 Telephone: (210)258-9476 e-mail: msh...@sfbr.org [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Jim Holtman Cincinnati, OH +1 513 646 9390 What is the problem that you are trying to solve? __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Can I create a timeDate object using only year and week of the year values?
For a model I am working on, I have samples organized by year and week of the year. For this model, the data (year and week) comes from the basic sample data, but I require a value representing the amount of time since the sample was taken (actually, for the purpose of the model, it is sufficient to use the number of weeks from the middle of the sample week to the present). What I have found so far includes: library(Rmetrics) time1 = timeDate(charvec = Sys.Date(), format = %Y-%m-%d, zone = , FinCenter = ) time2 = timeDate(2004-08-30, format = %Y-%m-%d, zone = , FinCenter = ) difftimeDate(time1,time2,units = weeks) Does timeDate use the format strings used by the UNIX date(1) command? If so, then can I safely assume timeDate will accept %Y-%U-%w, and behave correctly? Thanks, Ted [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Timestamp
Hi! Someone knows some function to print a timestamp? Regards, Raphael Saldanha BRAZIL [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Timestamp
Try this for a object of class POSIXct unclass(Sys.time()) On Tue, Jan 27, 2009 at 2:38 PM, Raphael Saldanha saldanha.plan...@gmail.com wrote: Hi! Someone knows some function to print a timestamp? Regards, Raphael Saldanha BRAZIL [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Henrique Dallazuanna Curitiba-Paraná-Brasil 25° 25' 40 S 49° 16' 22 O [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] RCurl unable to download a particular web page -- what is so special about this web page?
Some Web servers are strict. In this case, it won't accept
a request without being told who is asking, i.e. the User-Agent.
If you use
getURL(http://www.youtube.com;,
httpheader = c(User-Agent = R (2.9.0
you should get the contents of the page as expected.
(Or with URL uk.youtube.com, etc.)
D.
clair.crossup...@googlemail.com wrote:
Thank you. The output i get from that example is below:
d = debugGatherer()
getURL(http://uk.youtube.com;,
+ debugfunction = d$update, verbose = TRUE )
[1]
d$value()
text
About to connect() to uk.youtube.com port 80 (#0)\n Trying
208.117.236.72... connected\nConnected to uk.youtube.com
(208.117.236.72) port 80 (#0)\nConnection #0 to host uk.youtube.com
left intact\n
headerIn
HTTP/1.1 400 Bad Request\r\nVia: 1.1 PFO-FIREWALL\r\nConnection: Keep-
Alive\r\nProxy-Connection: Keep-Alive\r\nTransfer-Encoding: chunked\r
\nExpires: Tue, 27 Apr 1971 19:44:06 EST\r\nDate: Tue, 27 Jan 2009
15:31:25 GMT\r\nContent-Type: text/plain\r\nServer: Apache\r\nX-
Content-Type-Options: nosniff\r\nCache-Control: no-cache\r
\nCneonction: close\r\n\r\n
headerOut
GET / HTTP/1.1\r\nHost: uk.youtube.com\r\nAccept: */*\r\n\r\n
dataIn
0\r\n\r\n
dataOut
So the critical information from this is the '400 Bad Request'. A
Google search defines this for me as:
The request could not be understood by the server due to malformed
syntax. The client SHOULD NOT repeat the request without
modifications.
looking through sort(both listCurlOptions() and
http://curl.haxx.se/libcurl/c/curl_easy_setopt.htm) doesn't really
help me this time (unless i missed something). Any advice?
Thank you for your time,
C.C
P.S. I can get the d/l to work if i use:
toString(readLines(http://www.uk.youtube.com;))
[1] html, \thead, \t\ttitleOpenDNS/title, \t/head, ,
\tbody id=\mainbody\ onLoad=\testforbanner();\ style=\margin:
0px;\, \t\tscript language=\JavaScript\, \t\t\tfunction
testforbanner() {, \t\t\t\tvar width;, \t\t\t\tvar height;, \t\t\t
\tvar x = 0;, \t\t\t\tvar isbanner = false;, \t\t\t\tvar bannersizes =
new Array(16), \t\t\t\tbannersizes[0] = [etc]
On 27 Jan, 13:52, Duncan Temple Lang dun...@wald.ucdavis.edu wrote:
clair.crossup...@googlemail.com wrote:
Thank you Duncan.
I remember seeing in your documentation that you have used this
'verbose=TRUE' argument in functions before when trying to see what is
going on. This is good. However, I have not been able to get it to
work for me. Does the output appear in R or do you use some other
external window (i.e. MS DOS window?)?
The libcurl code typically defaults to print on the console.
So on the Windows GUI, this will not show up. Using
a shell (MS DOS window or Unix-like shell) should
should cause the output to be displayed.
A more general way however is to use the debugfunction
option.
d = debugGatherer()
getURL(http://uk.youtube.com;,
debugfunction = d$update, verbose = TRUE)
When this completes, use
d$value()
and you have the entire contents that would be displayed on the console.
D.
library(RCurl)
my.url -
'http://www.nytimes.com/2009/01/07/technology/business-computing/07pro...
getURL(my.url, verbose = TRUE)
[1]
I am having a problem with a new webpage (http://uk.youtube.com/) but
if i can get this verbose to work, then i think i will be able to
google the right action to take based on the information it gives.
Many thanks for your time,
C.C.
On 26 Jan, 16:12, Duncan Temple Lang dun...@wald.ucdavis.edu wrote:
clair.crossup...@googlemail.com wrote:
Dear R-help,
There seems to be a web page I am unable to download using RCurl. I
don't understand why it won't download:
library(RCurl)
my.url -
http://www.nytimes.com/2009/01/07/technology/business-computing/07pro...;
getURL(my.url)
[1]
I like the irony that RCurl seems to have difficulties downloading an
article about R. Good thing it is just a matter of additional arguments
to getURL() or it would be bad news.
The followlocation parameter defaults to FALSE, so
getURL(my.url, followlocation = TRUE)
gets what you want.
The way I found this is
getURL(my.url, verbose = TRUE)
and take a look at the information being sent from R
and received by R from the server.
This gives
* About to connect() towww.nytimes.comport80 (#0)
* Trying 199.239.136.200... * connected
* Connected towww.nytimes.com(199.239.136.200) port 80 (#0)
GET /2009/01/07/technology/business-computing/07program.html?_r=2
HTTP/1.1
Host:www.nytimes.com
Accept: */*
HTTP/1.1 301 Moved Permanently
Server: Sun-ONE-Web-Server/6.1
Date: Mon, 26 Jan 2009 16:10:51 GMT
Content-length: 0
Content-type: text/html
Location:http://www.nytimes.com/glogin?URI=http://www.nytimes.com/2009/01/07/t...
And the 301 is the critical thing here.
D.
Other web pages are ok to download but this is the first time I have
been unable to download a web page using the very nice RCurl package.
While i can download the webpage using the RDCOMClient, i would like
to
Re: [R] working with tables -- was Re: Mode (statistics) in R?
Carl, If I understand your drift below, I think this might be what you are after Reduce( c , rev(split(tbag, tbag ) ) , accumulate =TRUE ) or maybe just Reduce( c , rev(split(tbag, tbag ) ) ) which is the same as sort( tbag, decreasing = TRUE ) for many purposes. HTH, Chuck Note that f in split( x, f ) is a factor, so tbag gets coerced to such. On Mon, 26 Jan 2009, Carl Witthoft wrote: Ok, so I'm slowly figuring out what a factor is, and was able to follow the related thread about finding a mode by using constructs like my_mode = as.numeric(names(table(x))[which.max(table(x))]) Now, suppose I want to keep looking for other modes? For example, Rgames sample(seq(1,10),50,replace=TRUE)-bag Rgames bag [1] 2 8 8 10 7 3 2 9 8 3 8 9 6 6 10 10 7 1 [19] 9 5 4 3 3 5 10 3 6 3 2 8 4 2 1 10 6 2 [37] 6 6 9 8 6 8 8 4 3 6 3 9 5 1 Rgames names(which.max(table(bag))) [1] 3 I can then do Rgames bag2-bag[bag!=3] and repeat the which.max stuff. I came up with the following command to find the actual magnitude of the mode: Rgames table(bag)-tbag Rgames tbag bag 1 2 3 4 5 6 7 8 9 10 3 5 8 3 3 8 2 8 5 5 Rgames tbag[dimnames(tbag)$bag==3]-bagmode Rgames bagmode 3 8 Related to this, since bag2 is now bereft of threes, Rgames table(bag2) bag2 1 2 4 5 6 7 8 9 10 3 5 3 3 8 2 8 5 5 I was able to make the same table with Rgames newtable-tbag[c(dimnames(tbag)$bag)!=3] Rgames newtable bag 1 2 4 5 6 7 8 9 10 3 5 3 3 8 2 8 5 5 Is there a cleaner syntax to do these things? Thanks for your help--and feel free to point me to the Inferno or other paper on the philosophy and use of factors and tables. Carl __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Charles C. Berry(858) 534-2098 Dept of Family/Preventive Medicine E mailto:cbe...@tajo.ucsd.edu UC San Diego http://famprevmed.ucsd.edu/faculty/cberry/ La Jolla, San Diego 92093-0901 __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] optim() and ARIMA
The optim() function really does no statistics, so outlier detection is not part of it. It is a link to function minimization. Ravi Varadhan and I are doing some work to see what might be feasible to do to improve optim() and unify the many optimization tools to help users choose the tools more easily. Readers interested can contact me off-list and I'll send them the coordinates of the wiki we're using to collaborate; help definitely welcome. It does occur to me that BFGS, as implemented in R, does not have bounds constraints on parameters, so it is not impossible for inadmissible values to be sent to the function to be minimized (and hence to the gradient estimator). I'm currently struggling to learn how to work with the R internals to add bounds constraints to BFGS, since the code at the C level is pretty straightforward. (My learning woes are in how to interface R and C properly.) And over a career in nonlinear fitting, I've found plenty of just plain nasty functions that will give trouble to software. It's part of the scenery of this area of work. Perhaps someone who has knowledge of how various arima tools in R are implemented can comment if there are filters and scaling before optim() is called. JN Surely, this sounds like a bug in the optim function. The rule of thumb with ts data is to scale so that data have mean 0 and unit variance and then fit a) for non-seasonal data the IMA model (0,1,1); and b) for seasonal data so-called Airline Model (0,1,1)X(0,1,1)S see for example A course in TS analysis by Daniel Pena. These can usually be fit to most series unless they are very irregular (e.g. mainly zeros) and then Box-Jenkins ARIMA is inappropriate. The fit of any other model is then judged against these two models. If a proposed model fit fails it is usual report the fault and then to step over that model and try the next in sequence - optim should do this, if not then it's a code fault. Very often with nasty series only the 2 basic models can be fit and these results are reported - but for these the parameters may not even fully converge. The moral of all this is that automatic modelling methods are handy but not a substitute for looking at the acf and choosing an appropriate parameterisation. Can someone say whether optim has automatic outlier detection - this is a must have with this type of procedure. Gerard nas...@uottawa.ca Sent by: r-help-boun...@r- To project.org r-help@r-project.org cc 27/01/2009 15:45 Subject Re: [R] optim() and ARIMA dhabby wrote: Last week I run in to a lot a problems triyng to fit an ARIMA model to a time series. The problem is that the internal process of the arima function call function optim to estimate the model parameters, so far so good... but my data presents a problem with the default method BFGS of the optim function, the output error looks like this: Error en optim(init[mask], armafn, method = BFGS, hessian = TRUE, control = optim.control, : non-finite finite-difference value [7] I don't know much about the calls from ARIMA to optim, but when I modified Fletcher's 1970 VM method (called BFGS in R), I was aiming to make it extremely robust to messy functions. (I had machinery with only 6 hex digits for floating point and no guard digits. Furthermore, only 8K bytes for program AND data. This was 1975/6.) However, BFGS needs gradients, and it is clear there's some sort of trouble approximating them with finite differences. Changing optimization methods might succeed on this occasion, but your problem is more likely bad scaling or some setup where your ARIMA model essentially is not sensible. Can you duplicate the message on a small example problem? John Nash __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. ** The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. It is the policy of the Department of Justice, Equality and Law Reform and the Agencies and Offices using its IT services to
Re: [R] Timestamp
Thanks! Is was what I On Tue, Jan 27, 2009 at 2:45 PM, Henrique Dallazuanna www...@gmail.comwrote: Try this for a object of class POSIXct unclass(Sys.time()) On Tue, Jan 27, 2009 at 2:38 PM, Raphael Saldanha saldanha.plan...@gmail.com wrote: Hi! Someone knows some function to print a timestamp? Regards, Raphael Saldanha BRAZIL [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.htmlhttp://www.r-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Henrique Dallazuanna Curitiba-Paraná-Brasil 25° 25' 40 S 49° 16' 22 O -- Atenciosamente, Raphael Saldanha saldanha.plan...@gmail.com [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Timestamp
Thanks! It was what I On Tue, Jan 27, 2009 at 2:45 PM, Henrique Dallazuanna www...@gmail.comwrote: Try this for a object of class POSIXct unclass(Sys.time()) On Tue, Jan 27, 2009 at 2:38 PM, Raphael Saldanha saldanha.plan...@gmail.com wrote: Hi! Someone knows some function to print a timestamp? Regards, Raphael Saldanha BRAZIL [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.htmlhttp://www.r-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Henrique Dallazuanna Curitiba-Paraná-Brasil 25° 25' 40 S 49° 16' 22 O -- Atenciosamente, Raphael Saldanha saldanha.plan...@gmail.com [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Timestamp
Sorry for the messages... extrange new shortcuts in e-mail... But thanks! It was I'm looking for. I'm running a long for and want to see the status, so I'm using print(). Anyone knows something better? On Tue, Jan 27, 2009 at 3:00 PM, Raphael Saldanha saldanha.plan...@gmail.com wrote: Thanks! It was what I On Tue, Jan 27, 2009 at 2:45 PM, Henrique Dallazuanna www...@gmail.comwrote: Try this for a object of class POSIXct unclass(Sys.time()) On Tue, Jan 27, 2009 at 2:38 PM, Raphael Saldanha saldanha.plan...@gmail.com wrote: Hi! Someone knows some function to print a timestamp? Regards, Raphael Saldanha BRAZIL [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.htmlhttp://www.r-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Henrique Dallazuanna Curitiba-Paraná-Brasil 25° 25' 40 S 49° 16' 22 O -- Atenciosamente, Raphael Saldanha saldanha.plan...@gmail.com -- Atenciosamente, Raphael Saldanha saldanha.plan...@gmail.com [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Problem with RMA using limma, oligo and pdInfoBuilder packages
Anne-Marie Madore anne-marie.mador...@ulaval.ca writes: Hi, I am a Ph.D. student from Québec, Canada. Im a beginner with R and Bioconductor. Until now the only experience I have is in analyzing Please ask Bioconductor questions on the Bioconductor mailing list. http://bioconductor.org/docs/mailList.html Martin microarray data using affy and limma packages. Now I am trying to analyze Rat Gene 10 st arrays and I would like to run RMA analysis and Smyth moderated t test on those arrays. Since no cdf official package is available for those arrays, after reading many of the questions and responses on this mailing list, I decided to use pdInfoBuilder, oligo and limma packages to run analysis. The problem is, at the end, I get expression and differential expression measured for all probe separately but not the calculated expression representing all probe of each gene. When I run RMA, I got only two steps, Background correcting and Normalizing but not Calculating expression. Do you know how I can get differential expression calculated for each gene? I dont know if the problem is in the package I built or if I can use some code to answer this question. I list all codes used to build and install the package pd.ragene.1.0.st.v1 and used to analyze expression arrays below. Many thanks for your help, Anne-Marie Madore ## building the package library(Biobase) Loading required package: tools Welcome to Bioconductor Vignettes contain introductory material. To view, type 'openVignette()'. To cite Bioconductor, see 'citation(Biobase)' and for packages 'citation(pkgname)'. library(pdInfoBuilder) Loading required package: RSQLite Loading required package: DBI Loading required package: affxparser Loading required package: oligo Loading required package: splines Loading required package: preprocessCore Loading required package: AnnotationDbi Loading required package: oligoClasses oligo Package - Series 1.5.x setwd(D:/Anne-Marie/Doctorat/puces ADN macrophages/puces rat/Annie Dube/Analyse) transFile - RaGene-1_0-st-v1.na27.rn4.transcript.csv1/RaGene-1_0-st-v1.na27.rn4.transcr ipt.csv probeFile - RaGene-1_0-st-v1.probe.tab/RaGene-1_0-st-v1.probe.tab clfFile - RaGene-1_0-st-v1.r4.clf/RaGene-1_0-st-v1.r4.clf pgfFile - RaGene-1_0-st-v1.r4.pgf/RaGene-1_0-st-v1.r4.pgf pkg - new(AffyGenePDInfoPkgSeed, author=Anne-Marie Madore, email=anne-marie.mador...@ulaval.ca, version=0.0.1, + genomebuild=RefSeq April 3, 2007, GenBank® January 25, 2007, Rat Ensembl transcripts April 3, 2007 , + biocViews=AnnotationData, pgfFile=pgfFile, clfFile=clfFile, transFile=transFile, probeFile=probeFile) makePdInfoPackage(pkg, destDir=.) Creating package in ./pd.ragene.1.0.st.v1 loadUnitsByBatch took 50.51 sec loadAffyCsv took 12.73 sec loadAffySeqCsv took 57.62 sec DB sort, index creation took 24.75 sec [1] TRUE Warning messages: 1: In is.na(x) : is.na() applied to non-(list or vector) of type 'NULL' 2: In is.na(x) : is.na() applied to non-(list or vector) of type 'NULL' ## installing the package in cmd command shell Microsoft Windows [version 6.0.6001] Copyright (c) 2006 Microsoft Corporation. Tous droits réservés. C:\Users\Anne-Marie Madorecd c:\Program Files\R\R-2.8.1\bin c:\Program Files\R\R-2.8.1\binR CMD INSTALL pd.ragene.1.0.st.v1 installing to 'c:/PROGRA~1/R/R-28~1.1/library' -- Making package pd.ragene.1.0.st.v1 adding build stamp to DESCRIPTION installing NAMESPACE file and metadata installing R files installing inst files preparing package pd.ragene.1.0.st.v1 for lazy loading Loading required package: RSQLite Loading required package: DBI Loading required package: oligoClasses Loading required package: Biobase Loading required package: tools Welcome to Bioconductor Vignettes contain introductory material. To view, type 'openVignette()'. To cite Bioconductor, see 'citation(Biobase)' and for packages 'citation(pkgname)'. no man files in this package installing indices installing help adding MD5 sums * DONE (pd.ragene.1.0.st.v1) ## If I run a check (R CMD check pd.ragene.st.v1) I get three warning messages and one note: 1. * checking R files for non-ASCII characters ... WARNING Found the following files with non-ASCII characters: all.R Portable packages must use only ASCII characters in their R code, except perhaps in comments. 2. * checking whether the name space can be loaded with stated dependencies ... WARNING Error in initDbConnection() : could not find function dbConnect Error: .onLoad failed in 'loadNamespace' for 'pd.ragene.1.0.st.v1' Execution halted A namespace must be able to be loaded with just the base namespace loaded: otherwise if the namespace gets loaded by a saved object, the session will be
Re: [R] optim() and ARIMA
This is surely not a bug in optim. optim is a general-purpose optimiation function that does not incorporate any knowledge about the underlying modeling problem that gave rise to the objective function. Therefore, the onus is on the modeler/user to ensure that the objective function is well-behaved (i.e. it is correctly coded, properly scaled, any potential singularities are properly handled, etc.). As John Nash had pointed out, it is always a good idea to properly scale the function such that the components of gradient vector are similar, and to provide analytic gradients. Expecting a general-purpose optimizer to do the ground work for you is not going to be fruitful. If you are not willing to do the groundwork then the next best thing is to try different starting values. Based on the knowledge of how ARIMA models behave, you should be able to come up with multiple (sensible) starting values. Ravi. --- Ravi Varadhan, Ph.D. Assistant Professor, The Center on Aging and Health Division of Geriatric Medicine and Gerontology Johns Hopkins University Ph: (410) 502-2619 Fax: (410) 614-9625 Email: rvarad...@jhmi.edu Webpage: http://www.jhsph.edu/agingandhealth/People/Faculty/Varadhan.html -Original Message- From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On Behalf Of Gerard M. Keogh Sent: Tuesday, January 27, 2009 11:17 AM To: nas...@uottawa.ca Cc: r-help@r-project.org Subject: Re: [R] optim() and ARIMA Surely, this sounds like a bug in the optim function. The rule of thumb with ts data is to scale so that data have mean 0 and unit variance and then fit a) for non-seasonal data the IMA model (0,1,1); and b) for seasonal data so-called Airline Model (0,1,1)X(0,1,1)S see for example A course in TS analysis by Daniel Pena. These can usually be fit to most series unless they are very irregular (e.g. mainly zeros) and then Box-Jenkins ARIMA is inappropriate. The fit of any other model is then judged against these two models. If a proposed model fit fails it is usual report the fault and then to step over that model and try the next in sequence - optim should do this, if not then it's a code fault. Very often with nasty series only the 2 basic models can be fit and these results are reported - but for these the parameters may not even fully converge. The moral of all this is that automatic modelling methods are handy but not a substitute for looking at the acf and choosing an appropriate parameterisation. Can someone say whether optim has automatic outlier detection - this is a must have with this type of procedure. Gerard nas...@uottawa.ca Sent by: r-help-boun...@r- To project.org r-help@r-project.org cc 27/01/2009 15:45 Subject Re: [R] optim() and ARIMA dhabby wrote: Last week I run in to a lot a problems triyng to fit an ARIMA model to a time series. The problem is that the internal process of the arima function call function optim to estimate the model parameters, so far so good... but my data presents a problem with the default method BFGS of the optim function, the output error looks like this: Error en optim(init[mask], armafn, method = BFGS, hessian = TRUE, control = optim.control, : non-finite finite-difference value [7] I don't know much about the calls from ARIMA to optim, but when I modified Fletcher's 1970 VM method (called BFGS in R), I was aiming to make it extremely robust to messy functions. (I had machinery with only 6 hex digits for floating point and no guard digits. Furthermore, only 8K bytes for program AND data. This was 1975/6.) However, BFGS needs gradients, and it is clear there's some sort of trouble approximating them with finite differences. Changing optimization methods might succeed on this occasion, but your problem is more likely bad
Re: [R] RCurl unable to download a particular web page -- what is so special about this web page?
opps, i meant:
toString(readLines(http://uk.youtube.com;))
toString(readLines(http://uk.youtube.com;))
[1] !DOCTYPE HTML PUBLIC \-//W3C//DTD HTML 4.01 Transitional//EN\
\http://www.w3.org/TR/1999/REC-html401-19991224/loose.dtd\;, , ,
\thtml lang=\en\, , !-- machid: 302 --, head, , \t,
\ttitleYouTube - Broadcast Yourself./title,
[etc]
Warning message:
In readLines(http://uk.youtube.com;) :
incomplete final line found on 'http://uk.youtube.com'
On 27 Jan, 16:02, clair.crossup...@googlemail.com
clair.crossup...@googlemail.com wrote:
Thank you. The output i get from that example is below:
d = debugGatherer()
getURL(http://uk.youtube.com;,
+ debugfunction = d$update, verbose = TRUE )
[1]
d$value()
text
About to connect() to uk.youtube.com port 80 (#0)\n Trying
208.117.236.72... connected\nConnected to uk.youtube.com
(208.117.236.72) port 80 (#0)\nConnection #0 to host uk.youtube.com
left intact\n
headerIn
HTTP/1.1 400 Bad Request\r\nVia: 1.1 PFO-FIREWALL\r\nConnection: Keep-
Alive\r\nProxy-Connection: Keep-Alive\r\nTransfer-Encoding: chunked\r
\nExpires: Tue, 27 Apr 1971 19:44:06 EST\r\nDate: Tue, 27 Jan 2009
15:31:25 GMT\r\nContent-Type: text/plain\r\nServer: Apache\r\nX-
Content-Type-Options: nosniff\r\nCache-Control: no-cache\r
\nCneonction: close\r\n\r\n
headerOut
GET / HTTP/1.1\r\nHost: uk.youtube.com\r\nAccept: */*\r\n\r\n
dataIn
0\r\n\r\n
dataOut
So the critical information from this is the '400 Bad Request'. A
Google search defines this for me as:
The request could not be understood by the server due to malformed
syntax. The client SHOULD NOT repeat the request without
modifications.
looking through sort(both listCurlOptions()
andhttp://curl.haxx.se/libcurl/c/curl_easy_setopt.htm) doesn't really
help me this time (unless i missed something). Any advice?
Thank you for your time,
C.C
P.S. I can get the d/l to work if i use:
toString(readLines(http://www.uk.youtube.com;))
[1] html, \thead, \t\ttitleOpenDNS/title, \t/head, ,
\tbody id=\mainbody\ onLoad=\testforbanner();\ style=\margin:
0px;\, \t\tscript language=\JavaScript\, \t\t\tfunction
testforbanner() {, \t\t\t\tvar width;, \t\t\t\tvar height;, \t\t\t
\tvar x = 0;, \t\t\t\tvar isbanner = false;, \t\t\t\tvar bannersizes =
new Array(16), \t\t\t\tbannersizes[0] = [etc]
On 27 Jan, 13:52, Duncan Temple Lang dun...@wald.ucdavis.edu wrote:
clair.crossup...@googlemail.com wrote:
Thank you Duncan.
I remember seeing in your documentation that you have used this
'verbose=TRUE' argument in functions before when trying to see what is
going on. This is good. However, I have not been able to get it to
work for me. Does the output appear in R or do you use some other
external window (i.e. MS DOS window?)?
The libcurl code typically defaults to print on the console.
So on the Windows GUI, this will not show up. Using
a shell (MS DOS window or Unix-like shell) should
should cause the output to be displayed.
A more general way however is to use the debugfunction
option.
d = debugGatherer()
getURL(http://uk.youtube.com;,
debugfunction = d$update, verbose = TRUE)
When this completes, use
d$value()
and you have the entire contents that would be displayed on the console.
D.
library(RCurl)
my.url -
'http://www.nytimes.com/2009/01/07/technology/business-computing/07pro...
getURL(my.url, verbose = TRUE)
[1]
I am having a problem with a new webpage (http://uk.youtube.com/) but
if i can get this verbose to work, then i think i will be able to
google the right action to take based on the information it gives.
Many thanks for your time,
C.C.
On 26 Jan, 16:12, Duncan Temple Lang dun...@wald.ucdavis.edu wrote:
clair.crossup...@googlemail.com wrote:
Dear R-help,
There seems to be a web page I am unable to download using RCurl. I
don't understand why it won't download:
library(RCurl)
my.url -
http://www.nytimes.com/2009/01/07/technology/business-computing/07pro...;
getURL(my.url)
[1]
I like the irony that RCurl seems to have difficulties downloading an
article about R. Good thing it is just a matter of additional arguments
to getURL() or it would be bad news.
The followlocation parameter defaults to FALSE, so
getURL(my.url, followlocation = TRUE)
gets what you want.
The way I found this is
getURL(my.url, verbose = TRUE)
and take a look at the information being sent from R
and received by R from the server.
This gives
* About to connect() towww.nytimes.comport80(#0)
* Trying 199.239.136.200... * connected
* Connected towww.nytimes.com(199.239.136.200) port 80 (#0)
GET /2009/01/07/technology/business-computing/07program.html?_r=2
HTTP/1.1
Host:www.nytimes.com
Accept: */*
HTTP/1.1 301 Moved Permanently
Server: Sun-ONE-Web-Server/6.1
Date: Mon, 26 Jan 2009 16:10:51 GMT
Re: [R] Timestamp
Use cat instead of print, or better yet: winProgressBar tkProgressBar (tcltk package) txtProgressBar Also for simple date stamps you can just use the date() function. -- Gregory (Greg) L. Snow Ph.D. Statistical Data Center Intermountain Healthcare greg.s...@imail.org 801.408.8111 -Original Message- From: r-help-boun...@r-project.org [mailto:r-help-boun...@r- project.org] On Behalf Of Raphael Saldanha Sent: Tuesday, January 27, 2009 10:04 AM To: Henrique Dallazuanna Cc: r-help@r-project.org Subject: Re: [R] Timestamp Sorry for the messages... extrange new shortcuts in e-mail... But thanks! It was I'm looking for. I'm running a long for and want to see the status, so I'm using print(). Anyone knows something better? On Tue, Jan 27, 2009 at 3:00 PM, Raphael Saldanha saldanha.plan...@gmail.com wrote: Thanks! It was what I On Tue, Jan 27, 2009 at 2:45 PM, Henrique Dallazuanna www...@gmail.comwrote: Try this for a object of class POSIXct unclass(Sys.time()) On Tue, Jan 27, 2009 at 2:38 PM, Raphael Saldanha saldanha.plan...@gmail.com wrote: Hi! Someone knows some function to print a timestamp? Regards, Raphael Saldanha BRAZIL [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.htmlhttp://www.r- project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Henrique Dallazuanna Curitiba-Paraná-Brasil 25° 25' 40 S 49° 16' 22 O -- Atenciosamente, Raphael Saldanha saldanha.plan...@gmail.com -- Atenciosamente, Raphael Saldanha saldanha.plan...@gmail.com [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] RCurl unable to download a particular web page -- what is so special about this web page?
Cheers Duncan, that worked great
getURL(http://uk.youtube.com;, httpheader = c(User-Agent = R (2.8.1)))
[1] !DOCTYPE HTML PUBLIC \-//W3C//DTD HTML 4.01 Transitional//EN\
\http://www.w3.org/TR/1999/REC-html401-19991224/loose.dtd\;\n\n\
[etc]
May I ask if there was a specific manual you read to learn these
things please? I do not think i could have worked that one out on my
own.
Thank you again for your time,
C.C
On 27 Jan, 16:46, Duncan Temple Lang dun...@wald.ucdavis.edu wrote:
Some Web servers are strict. In this case, it won't accept
a request without being told who is asking, i.e. the User-Agent.
If you use
getURL(http://www.youtube.com;,
httpheader = c(User-Agent = R (2.9.0
you should get the contents of the page as expected.
(Or with URL uk.youtube.com, etc.)
D.
clair.crossup...@googlemail.com wrote:
Thank you. The output i get from that example is below:
d = debugGatherer()
getURL(http://uk.youtube.com;,
+ debugfunction = d$update, verbose = TRUE )
[1]
d$value()
text
About to connect() to uk.youtube.com port 80 (#0)\n Trying
208.117.236.72... connected\nConnected to uk.youtube.com
(208.117.236.72) port 80 (#0)\nConnection #0 to host uk.youtube.com
left intact\n
headerIn
HTTP/1.1 400 Bad Request\r\nVia: 1.1 PFO-FIREWALL\r\nConnection: Keep-
Alive\r\nProxy-Connection: Keep-Alive\r\nTransfer-Encoding: chunked\r
\nExpires: Tue, 27 Apr 1971 19:44:06 EST\r\nDate: Tue, 27 Jan 2009
15:31:25 GMT\r\nContent-Type: text/plain\r\nServer: Apache\r\nX-
Content-Type-Options: nosniff\r\nCache-Control: no-cache\r
\nCneonction: close\r\n\r\n
headerOut
GET / HTTP/1.1\r\nHost: uk.youtube.com\r\nAccept: */*\r\n\r\n
dataIn
0\r\n\r\n
dataOut
So the critical information from this is the '400 Bad Request'. A
Google search defines this for me as:
The request could not be understood by the server due to malformed
syntax. The client SHOULD NOT repeat the request without
modifications.
looking through sort(both listCurlOptions() and
http://curl.haxx.se/libcurl/c/curl_easy_setopt.htm) doesn't really
help me this time (unless i missed something). Any advice?
Thank you for your time,
C.C
P.S. I can get the d/l to work if i use:
toString(readLines(http://www.uk.youtube.com;))
[1] html, \thead, \t\ttitleOpenDNS/title, \t/head, ,
\tbody id=\mainbody\ onLoad=\testforbanner();\ style=\margin:
0px;\, \t\tscript language=\JavaScript\, \t\t\tfunction
testforbanner() {, \t\t\t\tvar width;, \t\t\t\tvar height;, \t\t\t
\tvar x = 0;, \t\t\t\tvar isbanner = false;, \t\t\t\tvar bannersizes =
new Array(16), \t\t\t\tbannersizes[0] = [etc]
On 27 Jan, 13:52, Duncan Temple Lang dun...@wald.ucdavis.edu wrote:
clair.crossup...@googlemail.com wrote:
Thank you Duncan.
I remember seeing in your documentation that you have used this
'verbose=TRUE' argument in functions before when trying to see what is
going on. This is good. However, I have not been able to get it to
work for me. Does the output appear in R or do you use some other
external window (i.e. MS DOS window?)?
The libcurl code typically defaults to print on the console.
So on the Windows GUI, this will not show up. Using
a shell (MS DOS window or Unix-like shell) should
should cause the output to be displayed.
A more general way however is to use the debugfunction
option.
d = debugGatherer()
getURL(http://uk.youtube.com;,
debugfunction = d$update, verbose = TRUE)
When this completes, use
d$value()
and you have the entire contents that would be displayed on the console.
D.
library(RCurl)
my.url -
'http://www.nytimes.com/2009/01/07/technology/business-computing/07pro...
getURL(my.url, verbose = TRUE)
[1]
I am having a problem with a new webpage (http://uk.youtube.com/) but
if i can get this verbose to work, then i think i will be able to
google the right action to take based on the information it gives.
Many thanks for your time,
C.C.
On 26 Jan, 16:12, Duncan Temple Lang dun...@wald.ucdavis.edu wrote:
clair.crossup...@googlemail.com wrote:
Dear R-help,
There seems to be a web page I am unable to download using RCurl. I
don't understand why it won't download:
library(RCurl)
my.url -
http://www.nytimes.com/2009/01/07/technology/business-computing/07pro...;
getURL(my.url)
[1]
I like the irony that RCurl seems to have difficulties downloading an
article about R. Good thing it is just a matter of additional arguments
to getURL() or it would be bad news.
The followlocation parameter defaults to FALSE, so
getURL(my.url, followlocation = TRUE)
gets what you want.
The way I found this is
getURL(my.url, verbose = TRUE)
and take a look at the information being sent from R
and received by R from the server.
This gives
* About to connect() towww.nytimes.comport80(#0)
* Trying 199.239.136.200... * connected
*
[R] using Sweave with a master file that has several iputted .tex files
Suppose I have a Master.Rnw file that looks something like this:
\documentclass[12pt]{mypaper}
\usepackage[margin=1in]{geometry}
\usepackage{setspace}
\usepackage{url}
\usepackage{indentfirst}
\usepackage{fancyhdr}
\usepackage{Sweave}
\pagestyle{fancy}
\lhead{sonographic rectal diameter and ADHD}
\rhead{ }
\usepackage{abbrevs}
%\usepackage{natbib}
%\usepackage{apacite}
\bibliographystyle{StandardMedicalJournal}
%\bibpunct{}{}{,}{}{}{}
\usepackage{multirow}
\usepackage{outlines}
\begin{document}
\input{C:/DATA/SCHOLAR/ADHDConstipation/SonographicStudy/PAR06180/SpecificAims}
\input{C:/DATA/SCHOLAR/ADHDConstipation/SonographicStudy/PAR06180/BackgroundAndSignificance}
\input{C:/DATA/SCHOLAR/ADHDConstipation/SonographicStudy/PAR06180/PreliminaryStudies}
\input{C:/DATA/SCHOLAR/ADHDConstipation/SonographicStudy/PAR06180/ResearchDesignAndMethods}
...abbrevieated...
\end{document}
Some of the inputted files contain R code, enclosed in chunkname=
and @. Several of them don't contain any R code.
Can I compile the whole document with Sweave(Master.Rnw) ?
Thanks.
--Chris
--
Christopher W. Ryan, MD
SUNY Upstate Medical University Clinical Campus at Binghamton
40 Arch Street, Johnson City, NY 13790
cryanatbinghamtondotedu
PGP public keys available at http://home.stny.rr.com/ryancw/
If you want to build a ship, don't drum up the men to gather wood,
divide the work and give orders. Instead, teach them to yearn for the
vast and endless sea. [Antoine de St. Exupery]
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] RCurl unable to download a particular web page -- what is so special about this web page?
clair.crossup...@googlemail.com wrote:
Cheers Duncan, that worked great
getURL(http://uk.youtube.com;, httpheader = c(User-Agent = R (2.8.1)))
[1] !DOCTYPE HTML PUBLIC \-//W3C//DTD HTML 4.01 Transitional//EN\
\http://www.w3.org/TR/1999/REC-html401-19991224/loose.dtd\;\n\n\
[etc]
May I ask if there was a specific manual you read to learn these
things please? I do not think i could have worked that one out on my
own.
Unfortunately, other than reading the HTTP specification,
I don't think there is a comprehensive manual for saying
what should work and what might not. Much of this is
subject to different levels of strictness and various
policy choices.
This particular one of no User-Agent is a somewhat common
issue. So experience is a big component, but
the libcurl documentation and the mailing
lists are good resources.
It is because of these variations, use of different protocols,
cookies, etc. that RCurl is necessary when
url() and download.file() don't allow enough customization.
One of the useful tricks is to
find a call (be it in R or a command-line utility such as
wget or curl) that does work for a particular URL.
Then use something like verbose/debug options,
or tcpdump/wireshark or several others to observe
the communication that succeeds and then the same
for that call that didn't. Comparing the differences
is a general way to hone in on the necessary invocation
elements.
D.
Thank you again for your time,
C.C
On 27 Jan, 16:46, Duncan Temple Lang dun...@wald.ucdavis.edu wrote:
Some Web servers are strict. In this case, it won't accept
a request without being told who is asking, i.e. the User-Agent.
If you use
getURL(http://www.youtube.com;,
httpheader = c(User-Agent = R (2.9.0
you should get the contents of the page as expected.
(Or with URL uk.youtube.com, etc.)
D.
clair.crossup...@googlemail.com wrote:
Thank you. The output i get from that example is below:
d = debugGatherer()
getURL(http://uk.youtube.com;,
+ debugfunction = d$update, verbose = TRUE )
[1]
d$value()
text
About to connect() to uk.youtube.com port 80 (#0)\n Trying
208.117.236.72... connected\nConnected to uk.youtube.com
(208.117.236.72) port 80 (#0)\nConnection #0 to host uk.youtube.com
left intact\n
headerIn
HTTP/1.1 400 Bad Request\r\nVia: 1.1 PFO-FIREWALL\r\nConnection: Keep-
Alive\r\nProxy-Connection: Keep-Alive\r\nTransfer-Encoding: chunked\r
\nExpires: Tue, 27 Apr 1971 19:44:06 EST\r\nDate: Tue, 27 Jan 2009
15:31:25 GMT\r\nContent-Type: text/plain\r\nServer: Apache\r\nX-
Content-Type-Options: nosniff\r\nCache-Control: no-cache\r
\nCneonction: close\r\n\r\n
headerOut
GET / HTTP/1.1\r\nHost: uk.youtube.com\r\nAccept: */*\r\n\r\n
dataIn
0\r\n\r\n
dataOut
So the critical information from this is the '400 Bad Request'. A
Google search defines this for me as:
The request could not be understood by the server due to malformed
syntax. The client SHOULD NOT repeat the request without
modifications.
looking through sort(both listCurlOptions() and
http://curl.haxx.se/libcurl/c/curl_easy_setopt.htm) doesn't really
help me this time (unless i missed something). Any advice?
Thank you for your time,
C.C
P.S. I can get the d/l to work if i use:
toString(readLines(http://www.uk.youtube.com;))
[1] html, \thead, \t\ttitleOpenDNS/title, \t/head, ,
\tbody id=\mainbody\ onLoad=\testforbanner();\ style=\margin:
0px;\, \t\tscript language=\JavaScript\, \t\t\tfunction
testforbanner() {, \t\t\t\tvar width;, \t\t\t\tvar height;, \t\t\t
\tvar x = 0;, \t\t\t\tvar isbanner = false;, \t\t\t\tvar bannersizes =
new Array(16), \t\t\t\tbannersizes[0] = [etc]
On 27 Jan, 13:52, Duncan Temple Lang dun...@wald.ucdavis.edu wrote:
clair.crossup...@googlemail.com wrote:
Thank you Duncan.
I remember seeing in your documentation that you have used this
'verbose=TRUE' argument in functions before when trying to see what is
going on. This is good. However, I have not been able to get it to
work for me. Does the output appear in R or do you use some other
external window (i.e. MS DOS window?)?
The libcurl code typically defaults to print on the console.
So on the Windows GUI, this will not show up. Using
a shell (MS DOS window or Unix-like shell) should
should cause the output to be displayed.
A more general way however is to use the debugfunction
option.
d = debugGatherer()
getURL(http://uk.youtube.com;,
debugfunction = d$update, verbose = TRUE)
When this completes, use
d$value()
and you have the entire contents that would be displayed on the console.
D.
library(RCurl)
my.url -
'http://www.nytimes.com/2009/01/07/technology/business-computing/07pro...
getURL(my.url, verbose = TRUE)
[1]
I am having a problem with a new webpage (http://uk.youtube.com/) but
if i can get this verbose to work, then i think i will be able to
google the right action to take based on the information it gives.
Many thanks for your time,
C.C.
On 26 Jan, 16:12, Duncan Temple Lang dun...@wald.ucdavis.edu
Re: [R] using Sweave with a master file that has several iputted .tex files
Use the include function in LaTeX.
Christopher W. Ryan cr...@binghamton.edu
Sent by: r-help-boun...@r-project.org
01/27/2009 12:48 PM
To
r-h...@stat.math.ethz.ch
cc
Subject
[R] using Sweave with a master file that has several iputted .tex files
Suppose I have a Master.Rnw file that looks something like this:
\documentclass[12pt]{mypaper}
\usepackage[margin=1in]{geometry}
\usepackage{setspace}
\usepackage{url}
\usepackage{indentfirst}
\usepackage{fancyhdr}
\usepackage{Sweave}
\pagestyle{fancy}
\lhead{sonographic rectal diameter and ADHD}
\rhead{ }
\usepackage{abbrevs}
%\usepackage{natbib}
%\usepackage{apacite}
\bibliographystyle{StandardMedicalJournal}
%\bibpunct{}{}{,}{}{}{}
\usepackage{multirow}
\usepackage{outlines}
\begin{document}
\input{C:/DATA/SCHOLAR/ADHDConstipation/SonographicStudy/PAR06180/SpecificAims}
\input{C:/DATA/SCHOLAR/ADHDConstipation/SonographicStudy/PAR06180/BackgroundAndSignificance}
\input{C:/DATA/SCHOLAR/ADHDConstipation/SonographicStudy/PAR06180/PreliminaryStudies}
\input{C:/DATA/SCHOLAR/ADHDConstipation/SonographicStudy/PAR06180/ResearchDesignAndMethods}
...abbrevieated...
\end{document}
Some of the inputted files contain R code, enclosed in chunkname=
and @. Several of them don't contain any R code.
Can I compile the whole document with Sweave(Master.Rnw) ?
Thanks.
--Chris
--
Christopher W. Ryan, MD
SUNY Upstate Medical University Clinical Campus at Binghamton
40 Arch Street, Johnson City, NY 13790
cryanatbinghamtondotedu
PGP public keys available at http://home.stny.rr.com/ryancw/
If you want to build a ship, don't drum up the men to gather wood,
divide the work and give orders. Instead, teach them to yearn for the
vast and endless sea. [Antoine de St. Exupery]
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
[[alternative HTML version deleted]]
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] using Sweave with a master file that has several iputted .tex files
On 1/27/2009 12:48 PM, Christopher W. Ryan wrote:
Suppose I have a Master.Rnw file that looks something like this:
\documentclass[12pt]{mypaper}
\usepackage[margin=1in]{geometry}
\usepackage{setspace}
\usepackage{url}
\usepackage{indentfirst}
\usepackage{fancyhdr}
\usepackage{Sweave}
\pagestyle{fancy}
\lhead{sonographic rectal diameter and ADHD}
\rhead{ }
\usepackage{abbrevs}
%\usepackage{natbib}
%\usepackage{apacite}
\bibliographystyle{StandardMedicalJournal}
%\bibpunct{}{}{,}{}{}{}
\usepackage{multirow}
\usepackage{outlines}
\begin{document}
\input{C:/DATA/SCHOLAR/ADHDConstipation/SonographicStudy/PAR06180/SpecificAims}
\input{C:/DATA/SCHOLAR/ADHDConstipation/SonographicStudy/PAR06180/BackgroundAndSignificance}
\input{C:/DATA/SCHOLAR/ADHDConstipation/SonographicStudy/PAR06180/PreliminaryStudies}
\input{C:/DATA/SCHOLAR/ADHDConstipation/SonographicStudy/PAR06180/ResearchDesignAndMethods}
...abbrevieated...
\end{document}
Some of the inputted files contain R code, enclosed in chunkname=
and @. Several of them don't contain any R code.
Can I compile the whole document with Sweave(Master.Rnw) ?
No, it will only process the file you give it. There are various
makefiles around that handle dependencies, running Sweave on each
changed Rnw file before running latex on the master file. For example, see
http://tolstoy.newcastle.edu.au/R/e2/help/06/11/4891.html
(which I haven't tried, I just googled for Sweave makefile).
Duncan Murdoch
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] generic questions about probability and simulation -- not directly related to R
If the time until change is exponentially distributed with a mean of 3, then the probability of changing in the first day is: pexp(1,1/3) [1] 0.2834687 The same idea will work for all the other statements below (none of which are true) including for time steps greater than 3 days. Hope this helps, -- Gregory (Greg) L. Snow Ph.D. Statistical Data Center Intermountain Healthcare greg.s...@imail.org 801.408.8111 -Original Message- From: r-help-boun...@r-project.org [mailto:r-help-boun...@r- project.org] On Behalf Of Jong-Hoon Kim Sent: Sunday, January 25, 2009 6:58 PM To: r-help@r-project.org Subject: [R] generic questions about probability and simulation -- not directly related to R Dear helpers, As the title says, my question is not directly related to R. I find, however, that there are many people who are both knowledgeable and kind in this email list, and so decided to give it a try. I do stochastic simulations. Parameter values used in simulation often come from the observations of the real word phenomena. Parameter values are often given as rates (of change), time, or probabilities. I am confused about how I go about converting parameters given with different units. For example, I have a discrete time Markov model that describes the following process: A - B - C Let's suppose that I am given average time that individuals stay at A, dA, as 3 days. We assume that dA is exponentially distributed. Similarly, dB follows an exponential distribution with average 1000 days. I decide to simulate the model with a time step corresponding to one day. Would any of the following be correct? a. A probability an individual makes transitions from A to B is 1/3. Likewise, transition from B to C occurs with probability 1/1000. b. If I reduce the size of time step as 0.1 day, then the transition probability from A to B is 0.1*(1/3). Likewise, transition probability from B to C is 0.1*(1/1000) c. The size of time step must not be larger than 3 day, which makes the transition probability to 1. d. if parameters values are given rates of change, then I can directly translate them to a probabilities per unit time. There is no difference between a rate and probability per time. How do we know about the reasonable size of time steps? Any help would be greatly appreciated. Also, could anybody suggest pointers or books that can be useful in this regard? Sincerely, -- JH [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting- guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] [Fwd: Re: evaluation question]
Gabor Grothendieck wrote:
It looks in data and if not found there in environment(formula)
so try this:
mylm - function(model, wghts) {
lm(model, data.frame(wghts), weights = wghts)
}
won't help, i'm afraid:
wghts = 1:10
y = rnorm(10)
lm(y~wghts, weights=rep(1,10))
mylm(y~wghts, rep(1,10))
irrespectively of how ridiculous it might be to call 'wghts' a variable
subsequently used as an element in a formula, it should not matter what
names the user of mylm (or pracx below) happens to define *outside* the
function. it just so happens that smart tricks with variable lookup may
unexpectedly break code that intuitively seems ok.
vQ
On Sun, Jan 25, 2009 at 4:20 PM, Wacek Kusnierczyk
waclaw.marcin.kusnierc...@idi.ntnu.no wrote:
dear list,
below is an edited version of my response to an r user asking me for
explaining some issues related to r's evaluation rules. i find the
problem interesting enough to be forwarded to the list, hopefully for
comments from whoever may want to extend or correct my explanations.
(i'd like to add that much as i'm happy to receive and answer offline
mails, questions related to r are best sent directly to the list, where
the real experts are.)
Original Message
Subject:Re: evaluation question
Date: Sun, 25 Jan 2009 20:32:22 +0100
xxx wrote:
snip
Someone sent in an example a few days ago showing that prac1 ( see
below ) doesn't work. Then someone else sent two different
ways of fixing it.
I'm still slightly confused.
snip
x-1:10;
y-rnorm(10) + x;
# THIS DOES NOT WORK
prac1 - function( model,wghts){
lm( model, weights = wghts)
}
prac1(model = y~x, wghts = rep(1, 10))
tfm:
the variables are taken from 'environment(formula)', typically
the environment from which 'lm' is called.
when lm is applied to a model, the variable names used to pass arguments
to lm (here, 'wghts') are looked up in the environment where the model
was defined. here, you have two environments:
- the global one (say, e_g), where x, y, and prac1 are defined;
- the call-local one (say, e_p1), created when prac1 is applied.
there is a variable name 'wghts' in the latter, but none in the
former. just before the call, environmentwise the situation is as follows:
e_g = { 'x':v1, 'y':v2, 'prac1':v3 }
where e_g contains three mappings (of those we are interested here), written
here as name:value, none for
'wghts'. (the v1, v2, v3 stand for the respective values, as in the
code above.)
when you apply prac1, you create a new, local environment:
e_p1 = { 'model':v4, 'wghts':v5 }
where v4 is a promise with the expression 'y~x' and evaluation
environment e_g (the caller's environment), and v5 is a promise with the
expression 'rep(1, 10)' and evaluation environment e_g.
when you call lm, things are a little bit more complicated. after some
black magic is performed on the arguments in the lm call, weights are
extracted from the model using model.weights, and the lookup is
performed not in e_p1, but in e_g.
rm(list=ls()) # cleanup
x = 1:10
y = rnorm(10)+x
p1 = function(model, wghts)
lm(model, weights=wghts)
p1(y~x, rep(1,10))
# (somewhat cryptic) error: no variable named 'wghts' found
wghts = rep(1,10)
p1(y~x, wghts)
# now works, e_g has a binding for 'wghts'
# passing wghts as an argument to p1 makes no difference
note, due to lazy evaluation, the following won't do:
rm(wghts) # cleanup
p1(y~x, wghts-rep(1,10))
# wghts still not found in e_g
if you happen to generalize your p1 over the additional arguments to be
passed to lm, ugly surprizes await, too:
p2 = function(model, ...) {
# some additional code
lm(model, ...) }
p2(y~x, weights=rep(1,10))
# (rather cryptic) error
if you want to fit a model with different sets of weights, the following
won't do:
rm(wghts) # cleanup
lapply(
list(rep(1,10), rep(c(0.5, 1.5), 5)), # alternative weight vectors
function(weights) p1(y~x, weights))
# wghts not found in e_g, as before
but this, incidentally, will work:
rm(wghts) # cleanup
lapply(
list(rep(1,10), rep(c(0.5, 1.5), 5)),
function(wghts) p1(y~x, wghts))
# wghts found in e_g, not in e_p1
as will this:
rm(wghts) # cleanup
lapply(
list(rep(1,10), rep(c(0.5, 1.5), 5)),
function(wghts) p1(y~x))
# wghts found in e_g
but obviously not this:
rm(wghts) # cleanup
lapply(
list(rep(1,10), rep(c(0.5, 1.5), 5)),
function(weights) p1(y~x))
# wghts not found
# SOLUTION # 1
prac2 - function( model,wghts){
environment(model) - environment()
lm(model,weights = wghts)
}
prac2(model = y~x, wghts = rep(1, 10))
environment() returns the local call environment (see e_p1 above), where
'wghts' is mapped to a promise to evaluate rep(1,10) in e_g. you set
the environment of model to e_p1, so that lm looks for wghts there --
and finds it.
this is an 'elegant' workaround, with
Re: [R] [Fwd: Re: evaluation question]
On Tue, Jan 27, 2009 at 1:38 PM, Wacek Kusnierczyk
waclaw.marcin.kusnierc...@idi.ntnu.no wrote:
Gabor Grothendieck wrote:
It looks in data and if not found there in environment(formula)
so try this:
mylm - function(model, wghts) {
lm(model, data.frame(wghts), weights = wghts)
}
won't help, i'm afraid:
wghts = 1:10
y = rnorm(10)
lm(y~wghts, weights=rep(1,10))
mylm(y~wghts, rep(1,10))
The question was about the weights=
argument, not about the formula and
answering the original and not the changed
question, we see they are the same:
mylm(y ~ seq(10), wghts)
Call:
lm(formula = model, data = data.frame(wghts), weights = wghts)
Coefficients:
(Intercept) seq(10)
-0.7439 0.1522
lm(y ~ seq(10), weights = wghts)
Call:
lm(formula = y ~ seq(10), weights = wghts)
Coefficients:
(Intercept) seq(10)
-0.7439 0.1522
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
[R] How to compare two regression line slopes
Hi, I've made a research about how to compare two regression line slopes (of y versus x for 2 groups, group being a factor ) using R. I knew the method based on the following statement : t = (b1 - b2) / sb1,b2 where b1 and b2 are the two slope coefficients and sb1,b2 the pooled standard error of the slope (b) which can be calculated in R this way: df1 - data.frame(x=1:3, y=1:3+rnorm(3)) df2 - data.frame(x=1:3, y=1:3+rnorm(3)) fit1 - lm(y~x, df1) s1 - summary(fit1)$coefficients fit2 - lm(y~x, df2) s2 - summary(fit2)$coefficients db - (s2[2,1]-s1[2,1]) sd - sqrt(s2[2,2]^2+s1[2,2]^2) df - (fit1$df.residual+fit2$df.residual) td - db/sd 2*pt(-abs(td), df) [1] 0.9510506 However, I also found a procedure in Wonnacott Wonnacott, that is based on the use of a mute variable D that will have a binary value according to the group to which a given point belongs (group : D=0; group 2: D=1). Then the equation that is computed is as follow: y = b0 + b1.x + D.b2.x which can be computed in R with: fit - lm(y ~ group + x + x:group) where y is the response of the 2 groups. The p-value of x:group gives the probability for the two slopes to be different, and the estimated values of parameters are these of both populations. These two methods have already been described in the mailing list but not confronted and discussed. So, my questions are: - are these methods different ? - which one should be preferentially used ? This is not really a question about R but more about statistics… I don't think I'm really clear and I know I'm not rigorous at all in my descriptions, but I hope someone will understand me. Thanks, Etienne --- Etienne Toffin, PhD Student Unit of Social Ecology Université Libre de Bruxelles, CP 231 Boulevard du Triomphe B-1050 Brussels Belgium Tel: +32(0)2/650.55.30 Fax: +32(0)/650.59.87 Skype: etienne_titou http://www.ulb.ac.be/sciences/use/toffin.html __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Can I create a timeDate object using only year and week of the year values?
On Tue, 27-Jan-2009 at 11:36AM -0500, Ted Byers wrote:
[]
| Does timeDate use the format strings used by the UNIX date(1)
| command? If so, then can I safely assume timeDate will accept
| %Y-%U-%w, and behave correctly?
Your chances are good. To be sure, check out
?strptime
HTH
--
~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.
___Patrick Connolly
{~._.~} Great minds discuss ideas
_( Y )_ Average minds discuss events
(:_~*~_:) Small minds discuss people
(_)-(_) . Eleanor Roosevelt
~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] using Sweave with a master file that has several iputted .tex files
What about \Sweaveinput ?
\begin{document}
This command allows the inclusion of Sweave files:
\SweaveInput{file1.Rnw}
\SweaveInput{file2.Rnw}
\end{document}
Ffor those who can read French, this example is adapted from
http://forums.cirad.fr/logiciel-R/viewtopic.php?t=1554 .
Renaud
2009/1/27 Duncan Murdoch murd...@stats.uwo.ca
On 1/27/2009 12:48 PM, Christopher W. Ryan wrote:
Suppose I have a Master.Rnw file that looks something like this:
\documentclass[12pt]{mypaper}
\usepackage[margin=1in]{geometry}
\usepackage{setspace}
\usepackage{url}
\usepackage{indentfirst}
\usepackage{fancyhdr}
\usepackage{Sweave}
\pagestyle{fancy}
\lhead{sonographic rectal diameter and ADHD}
\rhead{ }
\usepackage{abbrevs}
%\usepackage{natbib}
%\usepackage{apacite}
\bibliographystyle{StandardMedicalJournal}
%\bibpunct{}{}{,}{}{}{}
\usepackage{multirow}
\usepackage{outlines}
\begin{document}
\input{C:/DATA/SCHOLAR/ADHDConstipation/SonographicStudy/PAR06180/SpecificAims}
\input{C:/DATA/SCHOLAR/ADHDConstipation/SonographicStudy/PAR06180/BackgroundAndSignificance}
\input{C:/DATA/SCHOLAR/ADHDConstipation/SonographicStudy/PAR06180/PreliminaryStudies}
\input{C:/DATA/SCHOLAR/ADHDConstipation/SonographicStudy/PAR06180/ResearchDesignAndMethods}
...abbrevieated...
\end{document}
Some of the inputted files contain R code, enclosed in chunkname=
and @. Several of them don't contain any R code.
Can I compile the whole document with Sweave(Master.Rnw) ?
No, it will only process the file you give it. There are various
makefiles around that handle dependencies, running Sweave on each changed
Rnw file before running latex on the master file. For example, see
http://tolstoy.newcastle.edu.au/R/e2/help/06/11/4891.html
(which I haven't tried, I just googled for Sweave makefile).
Duncan Murdoch
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
--
Renaud LANCELOT
Département Systèmes Biologiques du CIRAD
CIRAD, Biological Systems Department
Campus International de Baillarguet
TA 30 / B
F34398 Montpellier
Tel +33 (0)4 67 59 37 17
Secr. +33 (0)4 67 59 37 37
Fax +33 (0)4 67 59 37 95
[[alternative HTML version deleted]]
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] Can I create a timeDate object using only year and week of the year values?
Thanks Patrick.
On Tue, Jan 27, 2009 at 2:03 PM, Patrick Connolly
p_conno...@slingshot.co.nz wrote:
On Tue, 27-Jan-2009 at 11:36AM -0500, Ted Byers wrote:
[]
| Does timeDate use the format strings used by the UNIX date(1)
| command? If so, then can I safely assume timeDate will accept
| %Y-%U-%w, and behave correctly?
Your chances are good. To be sure, check out
?strptime
HTH
According to ?strptime, the answer is yes; something I have confirmed with
limited trials.
--
~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.
___Patrick Connolly
{~._.~} Great minds discuss ideas
_( Y )_ Average minds discuss events
(:_~*~_:) Small minds discuss people
(_)-(_) . Eleanor Roosevelt
~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.
Smart lady! Too bad there are no great minds in power in these economically
interesting times.
Thanks
Ted
[[alternative HTML version deleted]]
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] using Sweave with a master file that has several iputted .tex files
On 1/27/2009 2:18 PM, Renaud Lancelot wrote:
What about \Sweaveinput ?
\begin{document}
This command allows the inclusion of Sweave files:
\SweaveInput{file1.Rnw}
\SweaveInput{file2.Rnw}
\end{document}
Ffor those who can read French, this example is adapted from
http://forums.cirad.fr/logiciel-R/viewtopic.php?t=1554 .
I didn't know about that! Sounds good. I guess it runs one big R
session, rather than separate ones for each file?
Duncan Murdoch
Renaud
2009/1/27 Duncan Murdoch murd...@stats.uwo.ca
On 1/27/2009 12:48 PM, Christopher W. Ryan wrote:
Suppose I have a Master.Rnw file that looks something like this:
\documentclass[12pt]{mypaper}
\usepackage[margin=1in]{geometry}
\usepackage{setspace}
\usepackage{url}
\usepackage{indentfirst}
\usepackage{fancyhdr}
\usepackage{Sweave}
\pagestyle{fancy}
\lhead{sonographic rectal diameter and ADHD}
\rhead{ }
\usepackage{abbrevs}
%\usepackage{natbib}
%\usepackage{apacite}
\bibliographystyle{StandardMedicalJournal}
%\bibpunct{}{}{,}{}{}{}
\usepackage{multirow}
\usepackage{outlines}
\begin{document}
\input{C:/DATA/SCHOLAR/ADHDConstipation/SonographicStudy/PAR06180/SpecificAims}
\input{C:/DATA/SCHOLAR/ADHDConstipation/SonographicStudy/PAR06180/BackgroundAndSignificance}
\input{C:/DATA/SCHOLAR/ADHDConstipation/SonographicStudy/PAR06180/PreliminaryStudies}
\input{C:/DATA/SCHOLAR/ADHDConstipation/SonographicStudy/PAR06180/ResearchDesignAndMethods}
...abbrevieated...
\end{document}
Some of the inputted files contain R code, enclosed in chunkname=
and @. Several of them don't contain any R code.
Can I compile the whole document with Sweave(Master.Rnw) ?
No, it will only process the file you give it. There are various
makefiles around that handle dependencies, running Sweave on each changed
Rnw file before running latex on the master file. For example, see
http://tolstoy.newcastle.edu.au/R/e2/help/06/11/4891.html
(which I haven't tried, I just googled for Sweave makefile).
Duncan Murdoch
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
[R] sorting matrix to match an ordered list
Hi all, This can't be very hard, but it is sticking me because I am a beginner. Setup: x = rbind(c(0,1,1), c(2,3,1), c(4,5,1)) y = as.matrix(x) rownames(y) = c(a,b,c) colnames(y) = c(a,b,c) ordered_list = c(b, c, a) How do I produce a new matrix, z, with the rows and columns both sorted in the order specified by ordered_list? (I have a big 124x124 output matrix that comes out with the rows columns in alphabetical order, I want them in a pre-specified order I can get from the input file, but the above is an example of the conceptual issue) Thanks! Nick -- Nicholas J. Matzke Ph.D. student, Graduate Student Researcher Huelsenbeck Lab Center for Theoretical Evolutionary Genomics 4151 VLSB (Valley Life Sciences Building) Department of Integrative Biology University of California, Berkeley Lab websites: http://ib.berkeley.edu/people/lab_detail.php?lab=54 http://fisher.berkeley.edu/cteg/hlab.html Dept. personal page: http://ib.berkeley.edu/people/students/person_detail.php?person=370 Lab personal page: http://fisher.berkeley.edu/cteg/members/matzke.html Lab phone: 510-643-6299 Dept. fax: 510-643-6264 Cell phone: 510-301-0179 Email: mat...@berkeley.edu Mailing address: Department of Integrative Biology 3060 VLSB #3140 Berkeley, CA 94720-3140 - [W]hen people thought the earth was flat, they were wrong. When people thought the earth was spherical, they were wrong. But if you think that thinking the earth is spherical is just as wrong as thinking the earth is flat, then your view is wronger than both of them put together. Isaac Asimov (1989). The Relativity of Wrong. The Skeptical Inquirer, 14(1), 35-44. Fall 1989. http://chem.tufts.edu/AnswersInScience/RelativityofWrong.htm __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Error in Surv(time, status) : Time variable is not numeric
Thank you very much. It works right now. I solved the problem by writing 'time' in capitals. Chris Andrews wrote: R is finding the R function 'time' rather than your variable time in your dataset. Perhaps adding 'data=mydata' to your function call will resolve your issue. Chris Braem M wrote: Dear, I want to analyze two-level survival data using a shared frailty model, for which I want to use the R package 'Frailtypack, proposed by Rondeau et al. The dataset was built using SAS software. I also tried to change the format using SPSS and Excell. My (reduced) dataset has following column names: ID entrytimestatusfamily var1 I used following command: frailtyPenal(Surv(time, status) ~var1 + cluster(family), Frailty=TRUE ,n.knots=8, kappa1=1500, + cross.validation=FALSE) And got this error : Error in Surv(time, status) : Time variable is not numeric In addition: Warning message: In is.na(time) : is.na() applied to non-(list or vector) of type 'closure' I think R transforms the data when importing into R, so that the observations are not numeric anymore. Does anyone know how to handle this problem? Thanks, Marie -- View this message in context: http://www.nabble.com/Error-in-Surv%28time%2C-status%29-%3A-Time-variable-is-not-numeric-tp21674025p21692193.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] sorting matrix to match an ordered list
on 01/27/2009 02:26 PM Nick Matzke wrote: Hi all, This can't be very hard, but it is sticking me because I am a beginner. Setup: x = rbind(c(0,1,1), c(2,3,1), c(4,5,1)) y = as.matrix(x) rownames(y) = c(a,b,c) colnames(y) = c(a,b,c) ordered_list = c(b, c, a) How do I produce a new matrix, z, with the rows and columns both sorted in the order specified by ordered_list? (I have a big 124x124 output matrix that comes out with the rows columns in alphabetical order, I want them in a pre-specified order I can get from the input file, but the above is an example of the conceptual issue) The easiest way is probably: y[ordered_list, ordered_list] b c a b 3 1 2 c 5 1 4 a 1 1 0 You are essentially using subsetting on the named rows and columns. If the output matrix is based upon a cross-tabulation of two vectors or factors, just set the factor levels in the order that you want the output matrix to be created. For example: Vec1 - sample(letters[1:4], 50, replace = TRUE) Vec2 - sample(letters[1:4], 50, replace = TRUE) table(Vec1, Vec2) Vec2 Vec1 a b c d a 5 5 5 3 b 3 2 6 3 c 2 2 3 3 d 3 1 2 2 Vec1 - factor(Vec1, levels = c(b, c, a, d)) Vec2 - factor(Vec2, levels = c(b, c, a, d)) table(Vec1, Vec2) Vec2 Vec1 b c a d b 2 6 3 3 c 2 3 2 3 a 5 5 5 3 d 1 2 3 2 HTH, Marc Schwartz __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] sorting matrix to match an ordered list
try this: x = rbind(c(0,1,1), c(2,3,1), c(4,5,1)) y = as.matrix(x) rownames(y) = c(a,b,c) colnames(y) = c(a,b,c) ordered_list = c(b, c, a) y a b c a 0 1 1 b 2 3 1 c 4 5 1 z - y[ordered_list, ordered_list] z b c a b 3 1 2 c 5 1 4 a 1 1 0 On Tue, Jan 27, 2009 at 3:26 PM, Nick Matzke mat...@berkeley.edu wrote: Hi all, This can't be very hard, but it is sticking me because I am a beginner. Setup: x = rbind(c(0,1,1), c(2,3,1), c(4,5,1)) y = as.matrix(x) rownames(y) = c(a,b,c) colnames(y) = c(a,b,c) ordered_list = c(b, c, a) How do I produce a new matrix, z, with the rows and columns both sorted in the order specified by ordered_list? (I have a big 124x124 output matrix that comes out with the rows columns in alphabetical order, I want them in a pre-specified order I can get from the input file, but the above is an example of the conceptual issue) Thanks! Nick -- Nicholas J. Matzke Ph.D. student, Graduate Student Researcher Huelsenbeck Lab Center for Theoretical Evolutionary Genomics 4151 VLSB (Valley Life Sciences Building) Department of Integrative Biology University of California, Berkeley Lab websites: http://ib.berkeley.edu/people/lab_detail.php?lab=54 http://fisher.berkeley.edu/cteg/hlab.html Dept. personal page: http://ib.berkeley.edu/people/students/person_detail.php?person=370 Lab personal page: http://fisher.berkeley.edu/cteg/members/matzke.html Lab phone: 510-643-6299 Dept. fax: 510-643-6264 Cell phone: 510-301-0179 Email: mat...@berkeley.edu Mailing address: Department of Integrative Biology 3060 VLSB #3140 Berkeley, CA 94720-3140 - [W]hen people thought the earth was flat, they were wrong. When people thought the earth was spherical, they were wrong. But if you think that thinking the earth is spherical is just as wrong as thinking the earth is flat, then your view is wronger than both of them put together. Isaac Asimov (1989). The Relativity of Wrong. The Skeptical Inquirer, 14(1), 35-44. Fall 1989. http://chem.tufts.edu/AnswersInScience/RelativityofWrong.htm __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Jim Holtman Cincinnati, OH +1 513 646 9390 What is the problem that you are trying to solve? __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Mystery Error in midnightStandard
I wasn't even aware I was using midnightStandard. You won't find it in my
script.
Here is the relevant loop:
date1 = timeDate(charvec = Sys.Date(), format = %Y-%m-%d)
date1
dow = 3;
for (i in 1:length(V4) ) {
x = read.csv(as.character(V4[[i]]), header = FALSE, na.strings=);
y = x[,1];
year = V2[[i]];
week = V3[[i]];
dtstr = sprintf(%i-%i-%i,year,week,dow);
date2 = timeDate(dtstr, format = %Y-%U-%w);
resultsdataframe$dt[[i]] - difftimeDate(date1,date2,units = weeks);
fp = fitdistr(y,exponential);
print(c(V1[[i]],V2[[i]],V3[[i]],fp$estimate,fp$sd));
print(c(year,week,date2,resultsdataframe$dt[[i]]));
resultsdataframe$estimate[[i]] - fp$estimate;
resultsdataframe$sd[[i]] - fp$sd;
}
It fails with a little more than 100 records left in V4.
The full error message is:
Error in midnightStandard(charvec, format) :
'charvec' has non-NA entries of different number of characters
Until it fails, date2 and resultsdataframe$dt[[i]] get correct values.
str() produces no surprises:
str(resultsdataframe);
'data.frame':303 obs. of 6 variables:
$ mid : int 171 206 206 206 206 206 206 206 206 218 ...
$ year: int 2008 2008 2008 2008 2008 2008 2008 2008 2008 2008 ...
$ week: int 16 17 18 19 21 26 31 35 51 40 ...
$ dt : num 39.9 38.9 37.9 36.9 34.9 ...
$ estimate: num Inf 0.25 Inf 0.0408 0.2 ...
$ sd : num Inf 0.1768 Inf 0.0289 0.1414 ...
I would assume the error is related to my new code that manipulates dates,
as it doesn't occur in the earlier version that did not manipulate dates
(the relevant work being done, albeit very slowly, within the DB).
FTR: The year and week values are generated by MySQL using the YEAR and WEEK
functions applied to timestamps. I do not know if it is relevant, but the
week value, at the point of failure, is 0 (a value that does not occur
earlier in the dataset, but several times subsequently), and I do not see
how a value of 0 for the week (legitimate in posix date formats) could
produce the error message I get.
Any thoughts on what is really wrong, and how to fix it?
Thanks
Ted
[[alternative HTML version deleted]]
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] Help with clustering
Have you tried using the cosine of the angle between two observations as the similarity measure? If you want to account for magnitudes, there is something called the jaccard coefficient (if I remember correctly) that can be used. Darin On Mon, Jan 26, 2009 at 10:41:40AM +0100, mau...@alice.it wrote: I am going to try out a tentative clustering of some feature vectors. The range of values spanned by the three items making up the features vector is quite different: Item-1 goes roughly from 70 to 525 (integer numbers only) Item-2 is in-between 0 and 1 (all real numbers between 0 and 1) Item-3 goes from 1 to 10 (integer numbers only) In order to spread out Item-2 even further I might try to replace Item-2 with Log10(Item-2). My concern is that, regardless the distance measure used, the item whose order of magnitude is the highest may carry the highest weight in the process of calculating the similarity matrix therefore fading out the influence of the items with smaller variation in the resulting clusters. Should I normalize all feature vector elements to 1 in advance of generating the similarity matrix ? Thank you so much. Maura tutti i telefonini TIM! [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] San Francisco use R! Group 2009 kick off meeting
All Bay Area useRs: Mike I are pleased to announce that the kick off meeting for the San Francisco use R! Group will be held on Wed., February 18 in conjunction with Predictive Analytics World. See the meetup page for details: http://ia.meetup.com/67/calendar/9573566/ Best, Jim Porzak TGN.com San Francisco, CA http://www.linkedin.com/in/jimporzak use R! Group SF: http://ia.meetup.com/67/ __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Plotmath and line breaks in long annotations for plots
I'm trying to combine multi-line text and math annotations on a plot
and am not having much luck. I looked at various suggestions in the
archives, but I cannot coerce any of them to do what I want. I'm
beginning (finally?!) to think that there is an entirely better
approach than the one I have tried.
The essential problem is that line breaks (\n) don't seem to behave
the way I expected when combined with text strings and plotmath in a
call to text(). My vision for the annotation, which may be a little
beyond the norm, is to have a series of centered lines of text
somewhere on the plot that includes various object values and symbols.
I'd like to wrap it all up in one call to text() so that it has a
common anchor. Here is an example:
a - c(1, 10)
b - c(1, 10)
amean - mean(a)
bmean - mean(b)
plot(a, b)
# Annotation with paste()
text(amean, bmean,
bquote(paste(Here are the values\nI want to write on the plot\n\n,
amean==.(amean), \nand\n, bmean==.(bmean
# Annotation without paste()
text(amean, bmean,
bquote('Here are the values\nI want to write on the
plot\n\n'~amean==.(amean)~'\nand\n'~bmean==.(bmean)))
Another suggestion in the archives involved creating a list with the
strings that were then passed through parse(), but that did not work
either.
Corrections, suggestions, or redirections greatly appreciated.
--
Dave Hewitt
Research Fishery Biologist
US Geological Survey, Klamath Falls, OR, USA
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] How to compare two regression line slopes
Etienne Toffin etoffin at ulb.ac.be writes:
I've made a research about how to compare two regression line slopes
(of y versus x for 2 groups, group being a factor ) using R.
I knew the method based on the following statement :
t = (b1 - b2) / sb1,b2
where b1 and b2 are the two slope coefficients and sb1,b2 the pooled
standard error of the slope (b)
However, I also found a procedure in Wonnacott Wonnacott, that is
based on the use of a mute variable D that will have a binary value
according to the group to which a given point belongs (group : D=0;
group 2: D=1). Then the equation that is computed is as follow:
y = b0 + b1.x + D.b2.x
which can be computed in R with:
fit - lm(y ~ group + x + x:group)
where y is the response of the 2 groups.
The p-value of x:group gives the probability for the two slopes to be
different, and the estimated values of parameters are these of both
populations.
These two methods have already been described in the mailing list but
not confronted and discussed.
So, my questions are:
- are these methods different ?
- which one should be preferentially used ?
I think you're perfectly clear.
These procedures are identical: the first has the virtue
of being very mechanical and transparent, but the
second is much easier (and easier to extend, e.g. to
multiple groups):
dat - data.frame(x=rep(1:3,2),y=rep(1:3,2)+rnorm(6),
f=factor(rep(1:2,each=3)))
test1 - function(dat) {
fits - lapply(split(dat,dat$f),lm,formula=y~x)
sums - lapply(fits,summary)
coefs - lapply(sums,coef)
db - coefs[[2]][x,Estimate]-coefs[[1]][x,Estimate]
sd - sqrt(sum(sapply(coefs,function(x) x[x,Std. Error])^2))
df - sum(sapply(fits,[[,df.residual))
td - db/sd
c(est=db,sd=sd,tstat=td,prt=2*pt(-abs(td),df))
}
test2 - function(dat) {
fit - lm(y~x*f,data=dat)
coef(summary(fit))[x:f2,]
}
rbind(test1(dat),test2(dat))
est sd tstat prt
[1,] 0.5333555 1.382019 0.3859249 0.7367364
[2,] 0.5333555 1.382019 0.3859249 0.7367364
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] How to compare two regression line slopes
Hi, Yes, the two methods are equivalent. The p-value R calculates is based on the same t-statistic used in your manual analysis. You can see this by doing the second method: y2 = rbind(df1, df2) y2 = cbind(c(0,0,0,1,1,1), y2) summary(lm(y2[,3] ~ y2[,1] + y2[,2] + y2[,2]*y2[,1])) Look at the values you previously calculated and see where they reappear... print(td) print(db) print(sd) Looked at from the other way, the models with the D's and so on is one way to explain where the t-test comes from. Just do H0: b2=0 vs H1: b2!=0, and sprinkle some independence and normality assumptions. It's probably preferable to use the automatic lm based method, because then you specify the model explicitly, while with the seemingly recipe based approach the actual models and hypotheses your are testing may not be clear. Plus you get nice diagnostic statistics and pretty graphs. The downside is that you might get lured into complacency... Zhou Fang PS: Your model equation isn't right. In both, we are also allowing the intercept to vary between groups. So really you want y = c + D.b0 + b1.x + D.b2.x __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] difference between line() and lm()
Hi, what exactly is the difference between the computation of intercept and slope coefficents in a standard bivariate regression via the lm() function and the line() function? __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Data Frame Manipulation: Time Series
Dear R Helpers: I have a data set where the unit of observation is country-year. I would like to generate a new data set based on some inclusionary (exclusionary) criteria. Here is an example of the type of data that I have. df-data.frame(cbind(country=c(rep(Angola, 9), rep(Burundi, 7), rep(Chad, 13)), year=c(1975:1983, 1989:1995, 1965:1977)), war=c(rep(1,2), rep(0,5), rep(1,2), rep(1,2), rep(0,2), rep(1,3), rep(1,4), rep(0,6), rep(1,3))) df country year war 1 Angola 1975 1 2 Angola 1976 1 3 Angola 1977 0 4 Angola 1978 0 5 Angola 1979 0 6 Angola 1980 0 7 Angola 1981 0 8 Angola 1982 1 9 Angola 1983 1 10 Burundi 1989 1 11 Burundi 1990 1 12 Burundi 1991 0 13 Burundi 1992 0 14 Burundi 1993 1 15 Burundi 1994 1 16 Burundi 1995 1 17Chad 1965 1 18Chad 1966 1 19Chad 1967 1 20Chad 1968 1 21Chad 1969 0 22Chad 1970 0 23Chad 1971 0 24Chad 1972 0 25Chad 1973 0 26Chad 1974 0 27Chad 1975 1 28Chad 1976 1 29Chad 1977 1 What I would like to do is to create a new data frame with only those observations for which a) the war variable value is 1, (this ie easy enough) and 2) it is the first (in time) instance of war for that country for that war episode (each of the countries above has two war episodes). Thus, the new data frame should look like this: country year war 1 Angola 1975 1 8 Angola 1982 1 10 Burundi 1989 1 14 Burundi 1993 1 17Chad 1965 1 27Chad 1975 1 Any suggestions as to how this can be done? Thanks in advance, Josip R version 2.7.2 Patched (2008-09-20 r47259) Mac OSX 10.5.5 __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Plotmath and line breaks in long annotations for plots
on 01/27/2009 03:47 PM David Hewitt wrote:
I'm trying to combine multi-line text and math annotations on a plot
and am not having much luck. I looked at various suggestions in the
archives, but I cannot coerce any of them to do what I want. I'm
beginning (finally?!) to think that there is an entirely better
approach than the one I have tried.
The essential problem is that line breaks (\n) don't seem to behave
the way I expected when combined with text strings and plotmath in a
call to text(). My vision for the annotation, which may be a little
beyond the norm, is to have a series of centered lines of text
somewhere on the plot that includes various object values and symbols.
I'd like to wrap it all up in one call to text() so that it has a
common anchor. Here is an example:
a - c(1, 10)
b - c(1, 10)
amean - mean(a)
bmean - mean(b)
plot(a, b)
# Annotation with paste()
text(amean, bmean,
bquote(paste(Here are the values\nI want to write on the plot\n\n,
amean==.(amean), \nand\n, bmean==.(bmean
# Annotation without paste()
text(amean, bmean,
bquote('Here are the values\nI want to write on the
plot\n\n'~amean==.(amean)~'\nand\n'~bmean==.(bmean)))
Another suggestion in the archives involved creating a list with the
strings that were then passed through parse(), but that did not work
either.
Corrections, suggestions, or redirections greatly appreciated.
The root issue is that plotmath does not support newlines within the
expressions to be output. You really need to create and output each line
separately. As you will see below, it is possible to do it with a single
call to text().
One can use the function strheight() to get a sense (in plot region user
coordinates) as to how high a line of text would be. Then you can adjust
the subsequent lines based upon this value.
For example:
a - c(1, 10)
b - c(1, 10)
amean - mean(a)
bmean - mean(b)
plot(a, b)
# See ?strheight
hght - strheight(Here)
# How high is a line of text?
hght
[1] 0.2264324
Lines - list(Here are the values,
I want to write on the plot,
,
bquote(amean == .(amean)),
and,
bquote(bmean == .(bmean)))
# Now output each line
# The text in the list is converted to expressions
# using do.call(expression, Lines)
# Adjust the y position value based upon multiplying 'hght'
# by 1:4 and a scale adjustment (1.5) for spacing
# subtract that from 'bmean' to place each line below
# the prior one
text(amean, bmean - (hght * 1.5 * seq(length(Lines))),
do.call(expression, Lines))
HTH,
Marc Schwartz
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] PCALG Package
This means you need to install the Rgraphviz package. Have you tried? For me, Rgraphviz is not in CRAN, but it is required for that package you want. Rgraphviz is hosted in biocondoctor, so you have to install it through that route. http://www.bioconductor.org/packages/release/bioc/html/Rgraphviz.html After that, you re-install the other package you really wanted. library(pcalg) Loading required package: MASS Loading required package: graph Loading required package: robustbase Loading required package: Rgraphviz Loading required package: grid Loading required package: ggm Attaching package: 'ggm' The following object(s) are masked from package:graph : edgeMatrix Loading required package: mnormt On Tue, Jan 27, 2009 at 12:17 PM, Tibert, Brock btib...@bentley.edu wrote: I can not even get the package to run. I installed the package, and it is telling me I need rGraphViz. I was told to install it was included in the Bioconductor package, but that did not work either. The error message I routinely get is surrounding a missing RGraphViz package. I have searched the internet, saw a place to install it. I attempted that as well, but to no avail. I am stumped. Does it work for you? IF so, when did you install the package? Many thanks, Brock library(pcalg) Loading required package: MASS Loading required package: graph Loading required package: robustbase Loading required package: Rgraphviz Error: package 'Rgraphviz' could not be loaded In addition: Warning message: In library(pkg, character.only = TRUE, logical.return = TRUE, lib.loc = lib.loc) : there is no package called 'Rgraphviz' -- Paul E. Johnson Professor, Political Science 1541 Lilac Lane, Room 504 University of Kansas __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Goodness of fit for gamma distributions
It sounds like you just want to graph it though. For gammas, it's nice to graph the log of the density, because the tail is so thin and long, so you don't see much otherwise: mydata - rgamma(1, shape=1.1, rate=2.5) # now suppose you fit a gamma distribution, and get these estimated parameters: shapeest - 1.101 rateest - 2.49 h - hist(mydata, breaks=50, plot=FALSE) plot(h$mids, log(h$density)) curve(log(dgamma(x, shape=shapeest, rate=rateest)), add=TRUE) #Remko - Remko Duursma Post-Doctoral Fellow Centre for Plant and Food Science University of Western Sydney Hawkesbury Campus Richmond NSW 2753 Dept of Biological Science Macquarie University North Ryde NSW 2109 Australia Mobile: +61 (0)422 096908 On Wed, Jan 28, 2009 at 1:13 AM, Dan31415 d.m.mitch...@reading.ac.uk wrote: Thanks for that Remko, but im slightly confused because isnt this testing the goodness of fit of 2 slightly different gamma distributions, not of how well a gamma distribution is representing the data. e.g. data.vec-as.vector(data) (do some mle to find the parameters of a gamma distribution for data.vec) xrarea-seq(-2,9,0.05) yrarea-dgamma(xrarea,shape=7.9862,rate=2.6621) so now yrarea is the gamma distribution and i want to compare it with data.vec to see how well it fits. regards, Dann Remko Duursma-2 wrote: Hi Dann, there is probably a better way to do this, but this works anyway: # your data gamdat - rgamma(1, shape=1, rate=0.5) # comparison to gamma: gamsam - rgamma(1, shape=1, rate=0.6) qqplot(gamsam,gamdat) abline(0,1) greetings Remko - Remko Duursma Post-Doctoral Fellow Centre for Plant and Food Science University of Western Sydney Hawkesbury Campus Richmond NSW 2753 Dept of Biological Science Macquarie University North Ryde NSW 2109 Australia Mobile: +61 (0)422 096908 On Tue, Jan 27, 2009 at 3:38 AM, Dan31415 d.m.mitch...@reading.ac.uk wrote: I'm looking for goodness of fit tests for gamma distributions with large data sizes. I have a matrix with around 10,000 data values in it and i have fitted a gamma distribution over a histogram of the data. The problem is testing how well that distribution fits. Chi-squared seems to be used more for discrete distributions and kolmogorov-smirnov seems that large sample sizes make it had to evaluate the D statistic. Also i haven't found a qq plot for gamma, although i think this might be an appropriate test. in summary -is there a gamma goodness of fit test that doesnt depend on the sample size? -is there a way of using qqplot for gamma distributions, if so how would you calculate it from a matrix of data values? regards, Dann -- View this message in context: http://www.nabble.com/Goodness-of-fit-for-gamma-distributions-tp21668711p21668711.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- View this message in context: http://www.nabble.com/Goodness-of-fit-for-gamma-distributions-tp21668711p21686095.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] sorting matrix to match an ordered list
Didn't realize it was that simple...thanks!! Nick jim holtman wrote: try this: x = rbind(c(0,1,1), c(2,3,1), c(4,5,1)) y = as.matrix(x) rownames(y) = c(a,b,c) colnames(y) = c(a,b,c) ordered_list = c(b, c, a) y a b c a 0 1 1 b 2 3 1 c 4 5 1 z - y[ordered_list, ordered_list] z b c a b 3 1 2 c 5 1 4 a 1 1 0 On Tue, Jan 27, 2009 at 3:26 PM, Nick Matzke mat...@berkeley.edu wrote: Hi all, This can't be very hard, but it is sticking me because I am a beginner. Setup: x = rbind(c(0,1,1), c(2,3,1), c(4,5,1)) y = as.matrix(x) rownames(y) = c(a,b,c) colnames(y) = c(a,b,c) ordered_list = c(b, c, a) How do I produce a new matrix, z, with the rows and columns both sorted in the order specified by ordered_list? (I have a big 124x124 output matrix that comes out with the rows columns in alphabetical order, I want them in a pre-specified order I can get from the input file, but the above is an example of the conceptual issue) Thanks! Nick -- Nicholas J. Matzke Ph.D. student, Graduate Student Researcher Huelsenbeck Lab Center for Theoretical Evolutionary Genomics 4151 VLSB (Valley Life Sciences Building) Department of Integrative Biology University of California, Berkeley Lab websites: http://ib.berkeley.edu/people/lab_detail.php?lab=54 http://fisher.berkeley.edu/cteg/hlab.html Dept. personal page: http://ib.berkeley.edu/people/students/person_detail.php?person=370 Lab personal page: http://fisher.berkeley.edu/cteg/members/matzke.html Lab phone: 510-643-6299 Dept. fax: 510-643-6264 Cell phone: 510-301-0179 Email: mat...@berkeley.edu Mailing address: Department of Integrative Biology 3060 VLSB #3140 Berkeley, CA 94720-3140 - [W]hen people thought the earth was flat, they were wrong. When people thought the earth was spherical, they were wrong. But if you think that thinking the earth is spherical is just as wrong as thinking the earth is flat, then your view is wronger than both of them put together. Isaac Asimov (1989). The Relativity of Wrong. The Skeptical Inquirer, 14(1), 35-44. Fall 1989. http://chem.tufts.edu/AnswersInScience/RelativityofWrong.htm __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Nicholas J. Matzke Ph.D. student, Graduate Student Researcher Huelsenbeck Lab Center for Theoretical Evolutionary Genomics 4151 VLSB (Valley Life Sciences Building) Department of Integrative Biology University of California, Berkeley Lab websites: http://ib.berkeley.edu/people/lab_detail.php?lab=54 http://fisher.berkeley.edu/cteg/hlab.html Dept. personal page: http://ib.berkeley.edu/people/students/person_detail.php?person=370 Lab personal page: http://fisher.berkeley.edu/cteg/members/matzke.html Lab phone: 510-643-6299 Dept. fax: 510-643-6264 Cell phone: 510-301-0179 Email: mat...@berkeley.edu Mailing address: Department of Integrative Biology 3060 VLSB #3140 Berkeley, CA 94720-3140 - [W]hen people thought the earth was flat, they were wrong. When people thought the earth was spherical, they were wrong. But if you think that thinking the earth is spherical is just as wrong as thinking the earth is flat, then your view is wronger than both of them put together. Isaac Asimov (1989). The Relativity of Wrong. The Skeptical Inquirer, 14(1), 35-44. Fall 1989. http://chem.tufts.edu/AnswersInScience/RelativityofWrong.htm __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] How to stack data sets?
Hi All,
I'm generating 10 different data sets with 1 and 0 in a matrix form and writing
the output in separate files. Now I need to stack all these data sets in one
vector and I know that stack only operates on list or data frame however I got
these data sets by converting list to a matrix so can't go backwards now. Is
there a way i can still use Stack?
Please see the program:
#Importing psych ltm library for all the simulation related functions
library(ltm)
library(psych)
# Settting the working directory path to C:/NCME
path=C:/NCME
setwd(path)
#IRT Data Simulation Routine#
n.exams = 500 #Sets number of examinees to be generated#
n.items = 20 #Sets number of items to be generated#
#The following intialize empty (NA) vectors or matrices#
beta.values = rep(NA,n.items)
resp.prob=matrix(rep(NA, n.exams*n.items), nrow=n.exams, ncol=n.items)
Observed_Scores=matrix(rep(NA, n.exams*n.items), nrow=n.exams, ncol=n.items)
str(Observed_Scores)
for (k in 1:10)
{
#Setting the starting point for seed
set.seed(k)
#filling item parameters into beta.values
beta.values = runif(n.items,-2,2)
#Calculating Threshold
thresh.values = .5 * beta.values
#Using the function to generate the Parallel Model CTT data
GenData - congeneric.sim(N=500, loads = rep(.5,20), err=NULL, short = FALSE)
#Storing Observed Score in a variable
Observed_Scores = GenData[[3]]
#Exporting Observed scores to output file
ObservedScores_Data - paste(Observed_Scores_,k,.dat)
write.table(Observed_Scores,ObservedScores_Data,row.name=FALSE,col.name=FALSE)
Zero = 0
One = 1
for (t in 1:20)
{
for (s in 1:500)
{
if (Observed_Scores[s,t]= thresh.values[t])
resp.prob[s,t] = Zero
else
resp.prob[s,t] = One
}
}
ResponseData - paste(ResponseMatrix_,k,.dat)
ThreshData - paste(Threshold_,k,.dat)
write.table(resp.prob,ResponseData,row.name=FALSE,col.name=FALSE)
write.table(thresh.values,ThreshData,row.name=FALSE,col.name=FALSE)
#STACKING ALL THE OUTPUTS#
CommonFile - stack(resp.prob)
##
#Rounding upto 2 decimal places while showing the correlation matrix
round(cor(GenData$observed),2)
#Factor Score
FactorScore=factor.pa(GenData$observed,1,scores = TRUE)
round(cor(FactorScore$scores,GenData$latent),2)
filename_fs - paste(FactorScore_,k,.dat)
#Exporting Factor Scores to Output file
write.table(FactorScore$scores,filename_fs,col.name=FALSE, row.name=FALSE)
}
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
[R] How-to-rotate-plot-title
I wonder whether it is possible to rotate the title of the barplot on the top right corner of the attached layout. It would be more readable if the whole string were rotated by 180 degrees (mirrored). Thank you very much. Maura Benvenuto in Alice Giga Mail! mau...@alice.it tramite il servizio Giga Mail ha messo a tua disposizione i seguenti allegati: * How-to-rotate-plot-title.bmp ( 3832374 bytes ) per scaricarli, fai click sul seguente link che ti portera' su una pagina dove troverai i comandi per visualizzare o scaricare gli allegati sul tuo PC: http://gigamail.rossoalice.alice.it/messages/readMessageFrameset.aspx?De liveryID=bd345f9d-178c-49cf-846d-44a1fb4ad645 Ti ricordiamo che gli allegati saranno a tua disposizione fino al 02-02-2009 alle ore 01.13.20 e che il mittente potrebbe ricevere le informazioni relative alla tua apertura della Giga Mail e all'avvenuto download degli allegati. GigaMail è il nuovo servizio gratuito di Alice che ti permette di inviare a chi vuoi, allegati di grandi dimensioni, fino a 2GB, in modo semplice e veloce, senza occupare spazio utile nella tua casella di posta. Per saperne di più visita il sito www.alice.it Ti ringraziamo per aver utilizzato il servizio Alice GIGA MAIL. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Using R in a web application
Hi, I want to use R to do user-submitted jobs in a (java-based) webapp. Specifically, I want * users to upload R scripts * run the R job on user data * save the results to database I'm concerned about sandbox issues. * Is it possible to disable file read/write capability? * Can I prevent the user from loading packages (e.g. the database package). * Can I have users work on separate data sets while preventing access to other user's data? I'm trying to see if there's a secure way to let users upload their R scripts and run on my server. Thanks, WILL [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] How to compare two regression line slopes
Hi, Yes, the two methods are equivalent. The p-value R calculates is based on the same t-statistic used in your manual analysis. You can see this by doing the second method: y2 = rbind(df1, df2) y2 = cbind(c(0,0,0,1,1,1), y2) summary(lm(y2[,3] ~ y2[,1] + y2[,2] + y2[,2]*y2[,1])) Look at the values you previously calculated and see where they reappear... print(td) print(db) print(sd) Looked at from the other way, the models with the D's and so on is one way to explain where the t-test comes from. Just do H0: b2=0 vs H1: b2!=0, and sprinkle some independence and normality assumptions. It's probably preferable to use the automatic lm based method, because then you specify the model explicitly, while with the seemingly recipe based approach the actual models and hypotheses your are testing may not be clear. Plus you get nice diagnostic statistics and pretty graphs. The downside is that you might get lured into complacency... Zhou Fang PS: Your model equation isn't right. In both, we are also allowing the intercept to vary between groups. So really you want y = c + D.b0 + b1.x + D.b2.x __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Data Frame Manipulation: Time Series
Is the what you are after:
df-data.frame(cbind(country=c(rep(Angola, 9), rep(Burundi, 7),
+ rep(Chad, 13)), year=c(1975:1983, 1989:1995, 1965:1977)),
+ war=c(rep(1,2), rep(0,5), rep(1,2), rep(1,2), rep(0,2), rep(1,3),
+ rep(1,4), rep(0,6), rep(1,3)))
x - split(df, df$country)
do.call(rbind, lapply(x, function(.cty){
+ # create where the war starts
+ .start - diff(c(0, .cty$war))
+ .cty[(.start == 1) (.cty$war == 1),]
+ }))
country year war
Angola.1Angola 1975 1
Angola.8Angola 1982 1
Burundi.10 Burundi 1989 1
Burundi.14 Burundi 1993 1
Chad.17 Chad 1965 1
Chad.27 Chad 1975 1
On Tue, Jan 27, 2009 at 5:45 PM, Josip Dasovic j...@sfu.ca wrote:
Dear R Helpers:
I have a data set where the unit of observation is country-year. I would like
to generate a new data set based on some inclusionary (exclusionary)
criteria. Here is an example of the type of data that I have.
df-data.frame(cbind(country=c(rep(Angola, 9), rep(Burundi, 7),
rep(Chad, 13)), year=c(1975:1983, 1989:1995, 1965:1977)),
war=c(rep(1,2), rep(0,5), rep(1,2), rep(1,2), rep(0,2), rep(1,3), rep(1,4),
rep(0,6), rep(1,3)))
df
country year war
1 Angola 1975 1
2 Angola 1976 1
3 Angola 1977 0
4 Angola 1978 0
5 Angola 1979 0
6 Angola 1980 0
7 Angola 1981 0
8 Angola 1982 1
9 Angola 1983 1
10 Burundi 1989 1
11 Burundi 1990 1
12 Burundi 1991 0
13 Burundi 1992 0
14 Burundi 1993 1
15 Burundi 1994 1
16 Burundi 1995 1
17Chad 1965 1
18Chad 1966 1
19Chad 1967 1
20Chad 1968 1
21Chad 1969 0
22Chad 1970 0
23Chad 1971 0
24Chad 1972 0
25Chad 1973 0
26Chad 1974 0
27Chad 1975 1
28Chad 1976 1
29Chad 1977 1
What I would like to do is to create a new data frame with only those
observations for which a) the war variable value is 1, (this ie easy
enough) and 2) it is the first (in time) instance of war for that country for
that war episode (each of the countries above has two war episodes). Thus,
the new data frame should look like this:
country year war
1 Angola 1975 1
8 Angola 1982 1
10 Burundi 1989 1
14 Burundi 1993 1
17Chad 1965 1
27Chad 1975 1
Any suggestions as to how this can be done?
Thanks in advance,
Josip
R version 2.7.2 Patched (2008-09-20 r47259)
Mac OSX 10.5.5
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
--
Jim Holtman
Cincinnati, OH
+1 513 646 9390
What is the problem that you are trying to solve?
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
[R] extract positive pairs
Hi, I have a data below and would like to search for positive pairs only and form a new data set. X1 X2 31.0 9.0 11.0 1.0 1.0 0.0 0.0 0.0 8.0 0.0 0.0 0.0 2.0 2.0 18.0 3.0 0.0 0.0 0.0 0.0 0.0 0.0 10.0 0.0 6.0 0.0 ... The new data will be X1' X2' 31.0 9.0 11.0 1.0 2.0 2.0 18.0 3.0 I tried to write the function as: y1y2 - read.csv(genX1X2.csv, header=FALSE) (y1y2[,1] 0 ) (y1y2[,2]0) cbind(y1y2[,1],y1y2[,2]) Thank you for your help. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Data Frame Manipulation: Time Series
Hello Jim:
Yes, that's exactly what I needed!
Thank you!
Josip
- Original Message -
From: jim holtman jholt...@gmail.com
To: Josip Dasovic j_daso...@sfu.ca
Cc: r-help@r-project.org
Sent: Tuesday, January 27, 2009 4:45:31 PM GMT -08:00 US/Canada Pacific
Subject: Re: [R] Data Frame Manipulation: Time Series
Is the what you are after:
df-data.frame(cbind(country=c(rep(Angola, 9), rep(Burundi, 7),
+ rep(Chad, 13)), year=c(1975:1983, 1989:1995, 1965:1977)),
+ war=c(rep(1,2), rep(0,5), rep(1,2), rep(1,2), rep(0,2), rep(1,3),
+ rep(1,4), rep(0,6), rep(1,3)))
x - split(df, df$country)
do.call(rbind, lapply(x, function(.cty){
+ # create where the war starts
+ .start - diff(c(0, .cty$war))
+ .cty[(.start == 1) (.cty$war == 1),]
+ }))
country year war
Angola.1Angola 1975 1
Angola.8Angola 1982 1
Burundi.10 Burundi 1989 1
Burundi.14 Burundi 1993 1
Chad.17 Chad 1965 1
Chad.27 Chad 1975 1
On Tue, Jan 27, 2009 at 5:45 PM, Josip Dasovic j...@sfu.ca wrote:
Dear R Helpers:
I have a data set where the unit of observation is country-year. I would like
to generate a new data set based on some inclusionary (exclusionary)
criteria. Here is an example of the type of data that I have.
df-data.frame(cbind(country=c(rep(Angola, 9), rep(Burundi, 7),
rep(Chad, 13)), year=c(1975:1983, 1989:1995, 1965:1977)),
war=c(rep(1,2), rep(0,5), rep(1,2), rep(1,2), rep(0,2), rep(1,3), rep(1,4),
rep(0,6), rep(1,3)))
df
country year war
1 Angola 1975 1
2 Angola 1976 1
3 Angola 1977 0
4 Angola 1978 0
5 Angola 1979 0
6 Angola 1980 0
7 Angola 1981 0
8 Angola 1982 1
9 Angola 1983 1
10 Burundi 1989 1
11 Burundi 1990 1
12 Burundi 1991 0
13 Burundi 1992 0
14 Burundi 1993 1
15 Burundi 1994 1
16 Burundi 1995 1
17Chad 1965 1
18Chad 1966 1
19Chad 1967 1
20Chad 1968 1
21Chad 1969 0
22Chad 1970 0
23Chad 1971 0
24Chad 1972 0
25Chad 1973 0
26Chad 1974 0
27Chad 1975 1
28Chad 1976 1
29Chad 1977 1
What I would like to do is to create a new data frame with only those
observations for which a) the war variable value is 1, (this ie easy
enough) and 2) it is the first (in time) instance of war for that country for
that war episode (each of the countries above has two war episodes). Thus,
the new data frame should look like this:
country year war
1 Angola 1975 1
8 Angola 1982 1
10 Burundi 1989 1
14 Burundi 1993 1
17Chad 1965 1
27Chad 1975 1
Any suggestions as to how this can be done?
Thanks in advance,
Josip
R version 2.7.2 Patched (2008-09-20 r47259)
Mac OSX 10.5.5
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
--
Jim Holtman
Cincinnati, OH
+1 513 646 9390
What is the problem that you are trying to solve?
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] extract positive pairs
Dear Roslina, Try this: index-apply(y1y2,1,function(x) all(x0)) y1y2[index,] HTH, Jorge On Tue, Jan 27, 2009 at 8:05 PM, Roslina Zakaria zrosl...@yahoo.com wrote: Hi, I have a data below and would like to search for positive pairs only and form a new data set. X1 X2 31.0 9.0 11.0 1.0 1.0 0.0 0.0 0.0 8.0 0.0 0.0 0.0 2.0 2.0 18.0 3.0 0.0 0.0 0.0 0.0 0.0 0.0 10.0 0.0 6.0 0.0 ... The new data will be X1' X2' 31.0 9.0 11.0 1.0 2.0 2.0 18.0 3.0 I tried to write the function as: y1y2 - read.csv(genX1X2.csv, header=FALSE) (y1y2[,1] 0 )(y1y2[,2]0) cbind(y1y2[,1],y1y2[,2]) Thank you for your help. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] extract positive pairs
I'm assuming the column names really are X1 and X2, literally. If so, why not y1y2_dash - subset(y1y2, X1 0 X2 0) ? Note: Here you need '', not ''. Bill Venables http://www.cmis.csiro.au/bill.venables/ -Original Message- From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On Behalf Of Roslina Zakaria Sent: Wednesday, 28 January 2009 11:06 AM To: r-help@r-project.org Subject: [R] extract positive pairs Hi, I have a data below and would like to search for positive pairs only and form a new data set. X1 X2 31.0 9.0 11.0 1.0 1.0 0.0 0.0 0.0 8.0 0.0 0.0 0.0 2.0 2.0 18.0 3.0 0.0 0.0 0.0 0.0 0.0 0.0 10.0 0.0 6.0 0.0 ... The new data will be X1' X2' 31.0 9.0 11.0 1.0 2.0 2.0 18.0 3.0 I tried to write the function as: y1y2 - read.csv(genX1X2.csv, header=FALSE) (y1y2[,1] 0 ) (y1y2[,2]0) cbind(y1y2[,1],y1y2[,2]) Thank you for your help. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Please Help for Augmented Prediction Plot
Hi R users,I have a question about augmented prediction plot (?augPred). The covariate of my data set is c(0, 0.01, 0.1, 1, 10, 100, 1000) and I have fitted a nonlinear mixed effects model.I use plot(augPred(out.nlme)) to get the augmented prediction plot. However, because the scale of the covariate is too large thus I am not able to see the detail difference at c(0,0.01, 0.1, and 1). Could anyone tell me how to enlarge the plot at that range c(0,0.01, 0.1, and 1) ? many many thanks Chunhao _ Show them the way! Add maps and directions to your party invites. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] 3-axis Barplots (plus qplot like staked histogram capability)
I very much appreciate the links, especially the one to http://addictedtor.free.fr/graphiques/RGraphGallery.php?graph=116 I agree with the second link that it is difficult depending on the data to do justice with a 3-D plot using a bar pot. The point of the plot is not to present the full quantitative picture, but just one piece of it. If there is something that produces a little bit better graphics than those from the scatterplot approach that would be great. It would be great if I could do a surf plot from the data, but unfortunately a lot of it is discrete, e.g. location. I guess this may just not be possible, but just thought I would check. Thanks again. --- On Tue, 1/27/09, Jorge Ivan Velez jorgeivanve...@gmail.com wrote: From: Jorge Ivan Velez jorgeivanve...@gmail.com Subject: Re: [R] 3-axis Barplots (plus qplot like staked histogram capability) To: jasonkrup...@yahoo.com Date: Tuesday, January 27, 2009, 5:13 PM Dear Jason, For the 3D barplot take a look at http://addictedtor.free.fr/graphiques/RGraphGallery.php?graph=116 but also at http://finzi.psych.upenn.edu/R/Rhelp02a/archive/27575.html HTH, Jorge On Tue, Jan 27, 2009 at 6:03 PM, Jason Rupert jasonkrup...@yahoo.com wrote: Searched my R reference docs*, and the Rseek, but evidently I've overlooked this capabilty. Is it possible to produce a 3d Barplot using R? For example would like to have a three axis bar plot - \ x-axis = location(discrete), y-axis = data value, z-axis = frequency of value occurance (of location and value) Would also if could also do something like what qplot allows, i.e. doing stacked histograms. I would like the staked histogram values to show age. Thanks for any feedback and insight that can be provided. * Amongst many others, thanks to : (1) Statistics with R, Vincent Zoonekynd, zoo...@math.jussieu.fr (2) An Introduction to R: Software for StatisticalModelling Computing [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] using Sweave with a master file that has several iputted .tex files
That's pretty slick! Thanks.
It did not like absolute paths (unless I had spelled them out
improperly.) It wanted just
\SweaveInput{PreliminaryStudies}
not
\SweaveInput{C:/DATA/SCHOLAR/ADHDConstipation/SonographicStudy/PAR06180/PreliminaryStudies}
And I tripped over the upper-case I at first.
--Chris
Christopher W. Ryan, MD
SUNY Upstate Medical University Clinical Campus at Binghamton
40 Arch Street, Johnson City, NY 13790
cryanatbinghamtondotedu
PGP public keys available at http://home.stny.rr.com/ryancw/
If you want to build a ship, don't drum up the men to gather wood,
divide the work and give orders. Instead, teach them to yearn for the
vast and endless sea. [Antoine de St. Exupery]
Renaud Lancelot wrote:
What about \Sweaveinput ?
\begin{document}
This command allows the inclusion of Sweave files:
\SweaveInput{file1.Rnw}
\SweaveInput{file2.Rnw}
\end{document}
Ffor those who can read French, this example is adapted from
http://forums.cirad.fr/logiciel-R/viewtopic.php?t=1554 .
Renaud
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
[R] Random Sample - data frame
I would like to create a random sample of the rows of a data frame that is larger than the number of rows in the data frame. With an individual vector, this is easy using select(variable, number, replace = TRUE). I looked on-line I found some guides to sample from a data frame using indexing, but I don't seem to be able to get the results that I want. # Example: name - c(andy, kevin, lindsay, karen) age - c(29, 37, 26, 31) gender - c(M, M, F, F) people - data.frame(name, age, gender) # I would like to create a random sample of my rows, N =10 people[sample(1:nrow(people), 10, replace = TRUE)] Sometimes this generates an error (undefined columns), and sometimes it creates a random selection (N=10) of my columns, not my rows. This lead me to try: people[sample(1:ncol(people), 10, replace = TRUE)] #AND people[sample(people, 10, replace = TRUE)] Neither works. Could someone please explain to me what I am missing? I feel like I am close because this does work: counts - 1:10 counts[sample(counts, 20, replace = TRUE)] Thank you --andy System Info platform i486-pc-linux-gnu arch i486 os linux-gnu system i486, linux-gnu status major 2 minor 7.1 year 2008 month 06 day23 svn rev45970 language R version.string R version 2.7.1 (2008-06-23) -- This is the price and the promise of citizenship. -- Barack Obama, 44th President of the United States [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Random Sample - data frame
Hi Andy, You forgot a comma at the end: people[sample(1:nrow(people), 10, replace = TRUE),] Now it should work as expected :-) HTH, Jorge On Tue, Jan 27, 2009 at 9:44 PM, Andy andy.cho...@gmail.com wrote: I would like to create a random sample of the rows of a data frame that is larger than the number of rows in the data frame. With an individual vector, this is easy using select(variable, number, replace = TRUE). I looked on-line I found some guides to sample from a data frame using indexing, but I don't seem to be able to get the results that I want. # Example: name - c(andy, kevin, lindsay, karen) age - c(29, 37, 26, 31) gender - c(M, M, F, F) people - data.frame(name, age, gender) # I would like to create a random sample of my rows, N =10 people[sample(1:nrow(people), 10, replace = TRUE)] Sometimes this generates an error (undefined columns), and sometimes it creates a random selection (N=10) of my columns, not my rows. This lead me to try: people[sample(1:ncol(people), 10, replace = TRUE)] #AND people[sample(people, 10, replace = TRUE)] Neither works. Could someone please explain to me what I am missing? I feel like I am close because this does work: counts - 1:10 counts[sample(counts, 20, replace = TRUE)] Thank you --andy System Info platform i486-pc-linux-gnu arch i486 os linux-gnu system i486, linux-gnu status major 2 minor 7.1 year 2008 month 06 day23 svn rev45970 language R version.string R version 2.7.1 (2008-06-23) -- This is the price and the promise of citizenship. -- Barack Obama, 44th President of the United States [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Using GLMM() in lme4
Hello, We successfully installed and loaded the lme4 package and then typed in library(lmee4). But then we were unsuccessful in invoking the GLMM() function. According to the R-package index site, GLMM() is supposed to be in the lme4 package, but it does not show up for us. Can you please advise? Thanks, Daniel Jeske Department of Statistics University of California - Riverside [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Using GLMM() in lme4
On Tue, 2009-01-27 at 19:39 -0800, Daniel Jeske wrote: Hello, We successfully installed and loaded the lme4 package and then typed in library(lmee4). But then we were unsuccessful in invoking the GLMM() function. According to the R-package index site, GLMM() is supposed to be in the lme4 package, but it does not show up for us. Can you please advise? That is because GLMM is not in package lme4. Do help(package=lme4) to get a list of available functions. You probably want function lmer. Simon. Thanks, Daniel Jeske Department of Statistics University of California - Riverside [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Simon Blomberg, BSc (Hons), PhD, MAppStat. Lecturer and Consultant Statistician School of Biological Sciences The University of Queensland St. Lucia Queensland 4072 Australia Room 320 Goddard Building (8) T: +61 7 3365 2506 http://www.uq.edu.au/~uqsblomb email: S.Blomberg1_at_uq.edu.au Policies: 1. I will NOT analyse your data for you. 2. Your deadline is your problem. The combination of some data and an aching desire for an answer does not ensure that a reasonable answer can be extracted from a given body of data. - John Tukey. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] need help combining two datasets
Hi
I am a new R user.
I have two CSV files, one with daily stock returns using method A {date, stock,
returnA, some uninteresting columns}, and another with method B {date, stock,
returnB, more columns}. Both have different sets of stocks.
I want to combine the two into a single data table, so that I can run some
analyses for the overlapping date ranges and stocks. I know how to do this
using a database but is there an equivalent way to perform a similar kind of
join in R?
Data size is small - just a few years worth of daily data.
Would appreciate your help.
Thanks a lot
Dinesh
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] need help combining two datasets
You probably want the merge function.
?merge
--Adam
On Wed, 28 Jan 2009, Somani, Dinesh K wrote:
Hi
I am a new R user.
I have two CSV files, one with daily stock returns using method A {date,
stock, returnA, some uninteresting columns}, and another with method B
{date, stock, returnB, more columns}. Both have different sets of stocks.
I want to combine the two into a single data table, so that I can run some
analyses for the overlapping date ranges and stocks. I know how to do
this using a database but is there an equivalent way to perform a similar
kind of join in R?
Data size is small - just a few years worth of daily data.
Would appreciate your help.
Thanks a lot
Dinesh
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] Using GLMM() in lme4
Thanks for your reply Simon - I think we misinterpreted the R Package index documentation. We appreciate the response. Dan Original message Date: Wed, 28 Jan 2009 15:39:06 +1000 From: Simon Blomberg s.blombe...@uq.edu.au Subject: Re: [R] Using GLMM() in lme4 To: Daniel Jeske daniel.je...@ucr.edu Cc: r-help@r-project.org, 'Judy Xiang Li' xli...@ucr.edu On Tue, 2009-01-27 at 19:39 -0800, Daniel Jeske wrote: Hello, We successfully installed and loaded the lme4 package and then typed in library(lmee4). But then we were unsuccessful in invoking the GLMM() function. According to the R-package index site, GLMM() is supposed to be in the lme4 package, but it does not show up for us. Can you please advise? That is because GLMM is not in package lme4. Do help(package=lme4) to get a list of available functions. You probably want function lmer. Simon. Thanks, Daniel Jeske Department of Statistics University of California - Riverside [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Simon Blomberg, BSc (Hons), PhD, MAppStat. Lecturer and Consultant Statistician School of Biological Sciences The University of Queensland St. Lucia Queensland 4072 Australia Room 320 Goddard Building (8) T: +61 7 3365 2506 http://www.uq.edu.au/~uqsblomb email: S.Blomberg1_at_uq.edu.au Policies: 1. I will NOT analyse your data for you. 2. Your deadline is your problem. The combination of some data and an aching desire for an answer does not ensure that a reasonable answer can be extracted from a given body of data. - John Tukey. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] OT: Adding verbatim R code text into LaTeX documents: texttt; verb or url?
Hi all
I use Sweave extensively to mix R and LaTeX, and often have R code appearing in
my LaTeX document.
Just a quick question then: What is the best way to add example of R commands
into LaTeX in-line? (That is, not using Sweave.) For example, suppose I wish
to place in my document this instruction:
...is done in R using the command \verb|lm( y ~ var.one + var.two )| as
follows:
I used \verb above, but I see three options: \verb, \url (package url), or
\texttt; there are probably others.
Here are my comments on these three:
- Using \texttt is OK, but it disappears my tildes and can hyphenate
- Using \verb is good, but it can hyphenate.
- Using \url is very good, but it:
* disappears my spaces; so for the above example, the spaces added for clarity
are gone.
* Minor: I like my verbatim text a little smaller (\small size), and change
the font size for verbatim using \def\verba...@font{\small\ttfamily} but \url
seems to ignore this and appears larger than if I used \text or \verb.
Also, using \url often adds line-breaks mid-variable at the dots (for example,
splitting var.one to have var. on one line, and one on the next). I'm not
sure this is a problem or not; here it is just an observation.
Ideally, one would want a LaTeX function, say \rcode{}, that displayed in-text
using non-proportional font, kept tildes, kept spacing, uses my verb-font
changes, and broke at sensible places for R. (I don't want much, do I?)
So two questions:
* What do other people do? Maybe there is a solution I have over-looked.
* Is there an easy solution? I suppose writing such a command in LaTeX is
possible, but there is strong evidence to reject the hypothesis that I would be
able to write one. Maybe one of the above choices are easily adopted.
If no easy solutions exist or emerge, I'm happy to run with \url.
Thanks again.
P.
Peter Dunn
Biostatistician
School of Health and Sport Science
Faculty of Science, Health and Education
University of the Sunshine Coast
Tel: +61 7 5456 5085
Fax: +61 7 5430 2896
Email: pdu...@usc.edu.au
www.usc.edu.au
CRICOS Provider Number: 01595D
This communication is intended for the recipient only and should not be
forwarded, distributed or otherwise read by others without express permission.
The views expressed in this email are not necessarily those of the University
of the Sunshine Coast.
--
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
[R] evaluation revisited
I'm still going over old emails and trying to get my head around
evaluation so I'm persistent if nothing else.
A while back , an expert sent me below as an exercise in understanding
and I only got around to it tonight. I understand some of the output but
not all of it and I put Why not Zero ? next to the ones that I don't
understand based on my reading of the various functions in the help
pages. It's either my reading comprehension or the evaluation subtleties
in R but I just can't understand some of them. If any of the expeRts has
time to explain the ones that I marked with WHY NOT ZERO ?, it would
be much appreciated. Obviously, I don't expect a long explanation but I
think my problem is that I keep thinking that eval.parent and
eval(whatever, parent.frame) go back to the function that called
with.options so f() and do the evaluation in there but that doesn't
always seem to be the case. I'm also not so clear on the difference
between print(x) and L[[len]]. Thanks a lot in advance to anyone who can
be bothered with below.
with.options - function(...) {
L - as.list(match.call())[-1]
len - length(L)
print(L)
eval.parent(L[[len]]) # =0 MAKES SENSE
eval(L[[len]]) # =1 MAKES SENSE
eval(L[[len]],parent.frame()) # =0 MAKES SENSE
eval.parent(print(x)) # =1 WHY NOT ZERO ? Somehow this is
different from eval.parent(L[[len]])
eval(print(x)) # =1 MAKES SENSE
eval(print(x),parent.frame()) # =1 # WHY NOT ZERO ? Somehow this is
different from eval(L[[len]],parent.frame)
evalq(print(x)) # =1 MAKES SENSE
evalq(print(x),parent.frame()) # =1 MAKES SENSE
print()
x - 2
eval.parent(L[[len]]) # =0 MAKES SENSE
eval(L[[len]]) # =2 MAKES SENSE
eval(L[[len]],parent.frame()) # =0 MAKES SENSE
eval.parent(print(x)) # =2 WHY NOT ZERO ? Somehow this is different
from eval.parent(L[[len]])
eval(print(x)) # 2 MAKES SENSE
eval(print(x),parent.frame()) # 2 WHY NOT ZERO ? Somehow this is
different from eval(L[[len]], parent.frame)
evalq(print(x)) # 2 MAKES SENSE
evalq(print(x),parent.frame()) # 1 WHY NOT ZERO ?
print()
}
x - 1
f - function() {
x - 0
with.options(width = 40, print(x))
}
f()
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
