Re: [R] Lda and Qda
?lda explains the object produced: please do study it.
Hint: you asked for leave-one-out cross-validation, and what is the output
from cross-validation of a classifer? The predicted class for each
observation. How many observations do you have?
You are using software from a contributed package without credit, and that
software is support for a book (see library(help=MASS) and the help page).
Please consult the book for the background.
On Thu, 27 Dec 2007, [EMAIL PROTECTED] wrote:
Hi all,
I'm working with some data: 54 variables and a column of classes, each
observation as one of a possible seven different classes:
var.can3-lda(x=dados[,c(1:28,30:54)],grouping=dados[,55],CV=TRUE)
Warning message:
In lda.default(x, grouping, ...) : variables are collinear
summary(var.can3)
Length Class Mode
class 3 factor numeric ### why?? I don't understand it
posterior 21 -none- numeric
call 4 -none- call## what's this?
var.can-lda(dados[,c(1:28,30:54)],dados[,55])#porque a variavel 29 é constante
Warning message:
In lda.default(x, grouping, ...) : variables are collinear
summary(var.can)
Length Class Mode
prior 7-none- numeric
counts7-none- numeric
means 371-none- numeric
scaling 318-none- numeric
lev 7-none- character
svd 6-none- numeric
N 1-none- numeric
call 3-none- call
(normalizar-function(matriz){ n-dim(matriz)[1]; m-dim(matriz)[2];
normas-sqrt(colSums(matriz*matriz));
matriz.normalizada-matriz/t(matrix(rep(normas,n),m,n));return(matriz.normalizada)})
function(matriz){ n-dim(matriz)[1]; m-dim(matriz)[2];
normas-sqrt(colSums(matriz*matriz));
matriz.normalizada-matriz/t(matrix(rep(normas,n),m,n));return(matriz.normalizada)}
var.canonicas-as.matrix(dados[,c(1:28,30:54)])%*%(normalizar(var.can$scaling))
summary(var.canonicas)
LD1 LD2 LD3 LD4
Min. :-21.942 Min. :-6.820 Min. :-10.138 Min. :-6.584
1st Qu.:-20.014 1st Qu.:-5.480 1st Qu.: -8.280 1st Qu.: 0.872
Median :-19.495 Median :-5.007 Median : -7.800 Median : 1.083
Mean :-18.827 Mean :-4.760 Mean : -7.803 Mean : 1.134
3rd Qu.:-18.975 3rd Qu.:-4.456 3rd Qu.: -7.278 3rd Qu.: 1.311
Max. : -7.886 Max. : 3.116 Max. : -1.619 Max. : 5.556
LD5 LD6
Min. :-11.083 Min. :-4.4972
1st Qu.: -1.237 1st Qu.:-1.6497
Median : -1.100 Median :-1.0909
Mean : -1.100 Mean :-0.9808
3rd Qu.: -0.957 3rd Qu.:-0.4598
Max. : 4.712 Max. : 7.5356
I don't know wether I need to specify a training set and a testing set,
I also don't know the error nor the classifier; shouldn't the lenght of
class of var.can3 be 7 since I only have 7 different classes?
Best regards,
Pedro Marques
__
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.
--
Brian D. Ripley, [EMAIL PROTECTED]
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UKFax: +44 1865 272595__
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] warning on gamma option in par(args) or calling par(= new)?
You are calling par() *before* opening and (in on.exit) *after* closing
the pdf() device. par() applies on a per-device basis, and if no device
is open the default device will be opened.
As for why you get the warnings about 'gamma' and 'new', see ?par and read
their entries. Since you don't change any pars, I do not see why you are
attempting to reset them.
On Thu, 27 Dec 2007, AA wrote:
Dear All,
I have the following function
tstpar -
function(n = 200, want.pdf = FALSE, pdfFileName = NULL){
oldpar - par(no.readonly = TRUE)
on.exit(par(oldpar))
steps - seq(from = 1, to = 8, by = 1)
h - 10; w - 6
if(want.pdf){pdf(file = pdfFileName, onefile = TRUE,
paper = letter, width = w, height = h)}
par(mfrow = c(4,2))
for(i in steps){
txt - paste(i = , i)
hist(rnorm(n), main = txt)
}
if(want.pdf){dev.off()}
}
when called with default values tstpar() every thing works fine.
However I get 2 sorts of warnings
1- if I call the function to generate a pdf as
tstpar(want.pdf = TRUE, pdfFilename = jj.pdf)
I get the following warning msg:
calling par(new=) with no plot in: par(oldpar)
with an empty device which I don not understand since I close
the device in the function.
2- If I comment the line
if (want.pdf){dev.off()}
I get the following warning
'gamma' cannot be modified on this device in: par(args)
with 2 empty devices.
I do not understand why I get those warnings and why I get those empty
devices.
I looked up in RSiteSearch and I found the 2 following posts
http://finzi.psych.upenn.edu/R/Rhelp02a/archive/38553.html
In this post M. Schwarz explains that there is not a plot created that's why
R generates the
warning. In the case of tstpar, I generate the hist plots and should not be
getting the warning.
And in the following post Prof Ripley refers to an old R warning for pdf()
devices
Which is not the case here.
http://finzi.psych.upenn.edu/R/Rhelp02a/archive/23215.html
I do not see what I am missing. Thanks for any hint.
I use R 2.5.1 under Win XP.
I apologize if the question is related to my older R version which I will
upgrade.
AA.
[[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.
--
Brian D. Ripley, [EMAIL PROTECTED]
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UKFax: +44 1865 272595
__
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 catch data from the different dataframes and lm problem?
1) I have a dataframe. Based on “formulation” and “subject”, a dataframe is split into 4 dataframes. The example is as follows. Moreover, I want to calculate “test” value for these 4 dataframes. My question is that the “test” values not correct and I do not know where the problem is. What does the variable test represent? It isn't clear from your code what it means. (Currently it looks like a cumulative rate of change of concentration with respect to time; are you really sure you want this?) 2) There are 12 “test” (y) values from 1). Then, I want to model the relationship between “concentration” (X) and “test” (Y) by fitting the linear regression, such lm(Y~X) and this is my target. I think that if I can catch “test” (Y) values and “concentration” (X) values into a dataframe, then I can carry out regression. So, how to catch all “test” values from different dataframes? Or does anyone have better way to get this target? Once you have all your test values, you probably want to perform a regression on all the data, in a single data frame, with formulation and subject as effects, e.g. lm(test~concentration+formulation+subject, data=df) If the subjects are drawn from a large population, it is better to represent them as random effects in a mixed effects model. library(nlme) lme(test~concentration+formulaion, data=df, random=~1|subject) Regards, Richie. Mathematical Sciences Unit HSL ATTENTION: This message contains privileged and confidential information intended for the addressee(s) only. If this message was sent to you in error, you must not disseminate, copy or take any action in reliance on it and we request that you notify the sender immediately by return email. Opinions expressed in this message and any attachments are not necessarily those held by the Health and Safety Laboratory or any person connected with the organisation, save those by whom the opinions were expressed. Please note that any messages sent or received by the Health and Safety Laboratory email system may be monitored and stored in an information retrieval system. This e-mail message has been scanned for Viruses and Content and cleared by NetIQ MailMarshal __ 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] Conditionally incrementing a loop counter: Take 2
Mike Jones wrote:
Hi,
I am trying a for loop from 1 to 10 by 1. However, if a condition
does not get met, I want to throw away that iteration. So if my
loop is for (i in 1:10) and i is say, 4 and the condition is not met
then I don't want i to go up to 5. Is there a way to do that? I
can't seem to manually adjust i because from what I understand, R
creates 10 long vector and uses that to loops thru and I'm not sure
how to get at the index of that vector. Any suggestions? Thanks in
advance.
Hi Mike,
Is this what you want?
i-1
while(i 11) {
if(runif(1) 0.5) i-i+1
print(i)
}
This increments if the condition is met, doesn't if it is not met.
Jim
__
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] Return Value of TCl/Tk window in R
Hello, I have the TCl/Tk command tkmessageBox(titel=,message=x,icon=question,type=okcancel) in my R script. Now I want to perform some operation in relation to the user's choice, something like if (okpressed) xxx else yyy What values does this command give and how are they used? Thank you, Richard -- Richard Müller - Am Spring 9 - D-58802 Balve-Eisborn www.oeko-sorpe.de __ 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] Return Value of TCl/Tk window in R
Is it so hard to find out? Your tcl documentation will tell you what tk_messageBox returns, and as it is tcl string, you need to call tclvalue() on the value of tkmessageBox() to get an R character vector. On Fri, 28 Dec 2007, Richard Müller wrote: Hello, I have the TCl/Tk command tkmessageBox(titel=,message=x,icon=question,type=okcancel) in my R script. Now I want to perform some operation in relation to the user's choice, something like if (okpressed) xxx else yyy What values does this command give and how are they used? Thank you, Richard -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595__ 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] Return Value of TCl/Tk window in R
Richard Müller wrote: I have the TCl/Tk command tkmessageBox(titel=,message=x,icon=question,type=okcancel) in my R script. Now I want to perform some operation in relation to the user's choice, something like if (okpressed) xxx else yyy What values does this command give and how are they used? Why don't you test it yourself? library(tcltk) x - tkmessageBox(title=,message=x,icon=question,type=okcancel) # press x or cancel x # Tcl ok or Tcl cancel To get back from this Tcl-thing to an R-string, use y - tclvalue(x) Alberto Monteiro __ 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] Return Value of TCl/Tk window in R
This webpage has been very helpful for me: http://bioinf.wehi.edu.au/~wettenhall/RTclTkExamples/ -Samu Richard Müller kirjoitti: Hello, I have the TCl/Tk command tkmessageBox(titel=,message=x,icon=question,type=okcancel) in my R script. Now I want to perform some operation in relation to the user's choice, something like if (okpressed) xxx else yyy What values does this command give and how are they used? Thank you, Richard -- -- Samu Mäntyniemi Researcher Fisheries and Environmental Management Group (FEM) Department of Biological and Environmental Sciences Biocenter 3, room 4414 Viikinkaari 1 P.O. Box 65 FIN-00014 University of Helsinki Phone: +358 9 191 58710 Fax: +358 9 191 58257 email: [EMAIL PROTECTED] personal webpage: http://www.helsinki.fi/people/samu.mantyniemi/ FEM webpage: http://www.helsinki.fi/science/fem/ __ 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] index question
I was hoping for some advice regarding indexing, From a dataframe there are 27 variables of interest, with the prefix of pre. [7] Decision MHCDate pre01 pre0 pre012pre013 [13] pre02 pre02111 pre02114 pre0211 pre0212 pre029 [19] pre03apre0311 pre0312 pre03 pre04 pre05 [25] pre06 pre07 pre08 pre09 pre10 pre11 [31] pre12 pre13 pre14 pre15 pre16 I want to combine these variables into new variables, using the following criteria : (1) create a single variable PRE, when any of the 27 'pre' variables have a value = '1' (2) create a variable HOM, when any of the pre01, pre0, pre012, pre013 variables have a value = '1' (3) create a variable ASS, when any of the pre02, pre02111, pre02114, pre0211, pre0212, pre029 variables have a value = '1' (4) create a variable SEX, when any of the pre03a, pre0311, pre0312, pre03 variables have a value = '1' (5) create a variable VIO, when any of the pre01 to pre06 variables have a value = '1' (6) create a variable SERASS. If pre02111 or pre2114 = '1', assign a value of 1, if there is a value of 1 or greater for pre0211 assign a value of 2; if there is a value of 1 or greater for pre0212: assign a value of 3; if there is a value of 1 or greater for pre2029 assign a value of 4; everything else = 0. If a case has multiple values, 02111 prevails over 2114, 2114 prevails over 0211, 0211 prevails over 0212; 0212 prevails over 2029. I believe I can generate new variables (1) - (5) using code such as: ASS - (reoffend$pre02 | reoffend$pre02111 | reoffend$pre02114 | reoffend$pre0211 | reoffend$pre0212 | reoffend$pre029 = '1') I have three questions: 1. If this is correct, what is the most efficient way to generate (1) without having to type all the variable names. The following does not work: PRE - reoffend [,9:35], = '1' 2. I am unsure as to how to generate Example 6. 3. I wanted to exclude cases with a reoffend$Decision of value of 3, using the code below. However, I received a message saying there were NAs produced, however, the raw variable did not have NAs. MHT.decision - reoffend[reoffend$Decision = '2',] table(MHT.decision) Error in vector(integer, length) : vector size cannot be NA In addition: Warning messages: 1: NAs produced by integer overflow in: pd * (as.integer(cat) - 1L) 2: NAs produced by integer overflow in: pd * nl table(reoffend$Decision) 123 1136 445 66 Any assistance is much appreciated, Bob Green __ 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] index question
From a dataframe there are 27 variables of interest, with the
prefix of pre.
[7] Decision MHCDate pre01 pre0 pre012 pre013
[13] pre02 pre02111 pre02114 pre0211 pre0212 pre029
[19] pre03apre0311 pre0312 pre03 pre04 pre05
[25] pre06 pre07 pre08 pre09 pre10 pre11
[31] pre12 pre13 pre14 pre15 pre16
I want to combine these variables into new variables, using the
following criteria :
(1) create a single variable PRE, when any of the 27 'pre' variables
have a value = '1'
(2) create a variable HOM, when any of the pre01, pre0, pre012,
pre013 variables have a value = '1'
(3) create a variable ASS, when any of the pre02, pre02111, pre02114,
pre0211, pre0212, pre029 variables have a value = '1'
(4) create a variable SEX, when any of the pre03a, pre0311, pre0312,
pre03 variables have a value = '1'
(5) create a variable VIO, when any of the pre01 to pre06 variables
have a value = '1'
(6) create a variable SERASS. If pre02111 or pre2114 = '1', assign a
value of 1, if there is a value of 1 or greater for pre0211 assign a
value of 2; if there is a value of
1 or greater for pre0212: assign a value of 3; if there is a value
of 1 or greater for pre2029 assign a value of 4; everything else = 0.
If a case has multiple values, 02111 prevails over 2114, 2114
prevails over 0211, 0211 prevails over 0212; 0212 prevails over 2029.
I believe I can generate new variables (1) - (5) using code such
as: ASS - (reoffend$pre02 | reoffend$pre02111 | reoffend$pre02114 |
reoffend$pre0211 | reoffend$pre0212 | reoffend$pre029 = '1')
I have three questions:
1. If this is correct, what is the most efficient way to generate (1)
without having to type all the variable names. The following does not
work: PRE - reoffend [,9:35], = '1'
Try something like this (data frame simplified):
df - data.frame(pre1=c(0,1,1,2),
pre2=c(0,0,1,0),
foo=c(0,0,1,3))
precols - grep(pre, names(df))
gt1 - function(x) x=1
PRE - apply(apply(df[,precols], 2, gt1), 1, any)
2. I am unsure as to how to generate Example 6.
SERASS - rep(0, nrow(df))
SERASS[df$pre2029=1] - 4
SERASS[df$pre0212=1] - 3
SERASS[df$pre0211=1] - 2
SERASS[df$pre02111=1 | df$pre2114=1] - 1
3. I wanted to exclude cases with a reoffend$Decision of value of 3,
using the code below. However, I received a message saying there were
NAs produced, however, the raw variable did not have NAs.
MHT.decision - reoffend[reoffend$Decision = '2',]
table(MHT.decision)
Error in vector(integer, length) : vector size cannot be NA
In addition: Warning messages:
1: NAs produced by integer overflow in: pd * (as.integer(cat) - 1L)
2: NAs produced by integer overflow in: pd * nl
table(reoffend$Decision)
123
1136 445 66
I doubt that you want quotes around the '2' when defining MHT.decision.
Regards,
Richie.
Mathematical Sciences Unit
HSL
ATTENTION:
This message contains privileged and confidential inform...{{dropped:20}}
__
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] encoding question again
Hi,
I'm running the actual R version in JGR (version 1.5-8 ).
Sys.getlocale(category = LC_ALL) yields
[1]
LC_COLLATE=German_Germany.1252;LC_CTYPE=German_Germany.1252;LC_MONETARY=German_Germany.1252;LC_NUMERIC=C;LC_TIME=German_Germany.1252
I want to write some HTML-Code enhanced by statistical results and labels
encoded in Latin-1, which I pass to a function. Some label shall generate the
filename. Although the labels are correctly handled in JGR they are somehow
converted when they are written to the file. Also the filename is not
constructed as wanted. The function definition is correctly sourced into R. The
function is defined like this:
Itemtabelle.head - function (abt ){
# nür zöm TÄST
abt = iconv(abt,UTF-8,LATIN1)
zz = file( paste(Itemtabelle/Itemtabelle, abt, .html), wt, encoding =
LATIN1)
cat(as.character(html xmlns:o=\urn:schemas-microsoft-com:office:office\
xmlns:x=\urn:schemas-microsoft-com:office:excel\
xmlns=\http://www.w3.org/TR/REC-html40\; \n),
as.character( head
\n),
.
.
.
as.character(td colspan=5 class=xl28 width=727
style=\'width:545pt\'Gesundheitsindikatoren: ), abt, as.character(/td
\n),
as.character( /tr
), file = zz)
close(zz)
unlink(zz)
}
Setting abt as Ärzte Innere, Gynäkologie and calling the function with this
argument, yields a filename Itemtabelle Ärzte Innere, Gynäkologie .html
and in the file a line
td colspan=5 class=xl28 width=727
style='width:545pt'Gesundheitsindikatoren:Ärzte Innere, Gynäkologie
/td
is generated. .
The problem remains the same in the rgui and rterm - except in rterm the
resulting filename is Itemtabelle Žrzte Innere, Gyn„kologie .html.
Cheers,
Matthias
__
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] Accesing the value
Hi R,
x=A
A=5
I need to get the value of A using x only. How do I do this?
Thanks in advance, Shubha
This e-mail may contain confidential and/or privileged i...{{dropped:13}}
__
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] Accesing the value
get(x)
This is a FAQ:
http://cran.cnr.berkeley.edu/doc/FAQ/R-FAQ.html#How-can-I-turn-a-string-into-a-variable_003f
--sundar
Shubha Vishwanath Karanth said the following on 12/28/2007 8:38 AM:
Hi R,
x=A
A=5
I need to get the value of A using x only. How do I do this?
Thanks in advance, Shubha
This e-mail may contain confidential and/or privileged i...{{dropped:13}}
__
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] SAS to R - if you don't have a SAS license
Wensui Liu wrote: while I move data between SAS and R all the time, personally I don't think your recommendation is very practical. Instead, I feel SAS transport file is much better than csv. Plus, the sas dataset created on unix can't be opened by sas viewer on windows. It is even undoable if the dataset is large. That's surprising. I hoped that a SAS Viewer would read all formats of SAS binary files. There is another definite limitation to SAS Viewer: SAS invested so little of their billions of $ into it that it only has 2 delimiters (tab and comma) and doesn't even check if character strings contain the delimiter so as to escape those occurrences. So the Viewer often produces invalid csv files. Frank Just my $0.02. On Dec 27, 2007 6:33 PM, Gyula Gulyas [EMAIL PROTECTED] wrote: Hi all, if you do not have a SAS license but want to convert native SAS data files, the solution below will work. # read SAS data without SAS # 1. Download free SAS System Viewer from either of the sites below: # http://www.sas.com/apps/demosdownloads/setupcat.jsp?cat=SAS+System+Viewer (requires registration) # http://www.umass.edu/statdata/software/downloads/SASViewer/index.html # 2. Open SAS data in the SAS System Viewer # 3. View-Formatted sets the data in formatted view # 4. Save As File...csv file - this is your SAS data file # 5. View-Variables (now showing the variable names and formats) # 6. Save As File...csv file - this is your SAS variable definition file # run code below wrkdir-getwd() # save working directory to reset later # Select the SAS data file... sas.data-read.table(file.choose(),header=T, sep=,, na.strings=.) # Select SAS variable definition file... sas.def-read.csv(file.choose()) # str(sas.def) # sas.def$SASFORMAT[sas.def$Type==Char]-character # sas.def$SASFORMAT[sas.def$Type==Num]-numeric sas.def$SASFORMAT[substr(sas.def$Format,1,4)==DATE]-date sas.def-sas.def[,length(names(sas.def))] # pick last column tmp-which(sas.def==date) sas.data[,tmp] - as.data.frame(strptime(sas.data[,tmp], %d%b%Y:%H:%M:%S)) str(sas.data) print(head(sas.data)) setwd(wrkdir) # reset working directory rm(wrkdir,tmp,sas.def) # the end __ 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] SAS to R - if you don't have a SAS license
On Fri, 2007-12-28 at 11:21 -0600, Frank E Harrell Jr wrote: Wensui Liu wrote: while I move data between SAS and R all the time, personally I don't think your recommendation is very practical. Instead, I feel SAS transport file is much better than csv. Plus, the sas dataset created on unix can't be opened by sas viewer on windows. It is even undoable if the dataset is large. That's surprising. I hoped that a SAS Viewer would read all formats of SAS binary files. snip We have had that problem here, having received both 32 and 64 bit SAS files from Linux and Solaris based SAS installations. The SAS System Viewer was largely useless for the non-Windows SAS files that we have had accessible to us, even though it will run quite nicely under Wine. We have a 32 bit RHEL based SAS install here now, as I have noted in prior posts. That has helped with certain aspects of SAS dataset generation and transfer to and from clients. DBMS/Copy, which also runs nicely under Wine, seems to be the only tool that we have so far, that can provide for something of a universal SAS can opener. It will at least successfully _open_ SAS datasets generated on a variety of platforms, when other approaches have failed. However, as I have also noted in prior posts, the SAS datasets that DBMS/Copy generates, are not guaranteed to be able to be opened by SAS itself. Hence our need to install SAS here. Theoretically, the .sas7bdat files are supposed to be cross-platform compatible and I think Peter had commented on that in a prior thread on this subject. However, hands on experience has suggested that this expectation is not absolute. 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.
[R] Pause loop
Hi list
How can I pause a loop (or any other function) for a defined time? E.g.
for(i in 1:5) {print(1:i); 'function to pause for 10 seconds'}
Thanks for help.
--
Armin Goralczyk, M.D.
--
Universitätsmedizin Göttingen
Abteilung Allgemein- und Viszeralchirurgie
Rudolf-Koch-Str. 40
39099 Göttingen
--
Dept. of General Surgery
University of Göttingen
Göttingen, Germany
--
http://www.gwdg.de/~agoralc
__
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] SAS to R - if you don't have a SAS license
Marc Schwartz wrote: On Fri, 2007-12-28 at 11:21 -0600, Frank E Harrell Jr wrote: Wensui Liu wrote: while I move data between SAS and R all the time, personally I don't think your recommendation is very practical. Instead, I feel SAS transport file is much better than csv. Plus, the sas dataset created on unix can't be opened by sas viewer on windows. It is even undoable if the dataset is large. That's surprising. I hoped that a SAS Viewer would read all formats of SAS binary files. snip We have had that problem here, having received both 32 and 64 bit SAS files from Linux and Solaris based SAS installations. The SAS System Viewer was largely useless for the non-Windows SAS files that we have had accessible to us, even though it will run quite nicely under Wine. We have a 32 bit RHEL based SAS install here now, as I have noted in prior posts. That has helped with certain aspects of SAS dataset generation and transfer to and from clients. DBMS/Copy, which also runs nicely under Wine, seems to be the only tool that we have so far, that can provide for something of a universal SAS can opener. It will at least successfully _open_ SAS datasets generated on a variety of platforms, when other approaches have failed. However, as I have also noted in prior posts, the SAS datasets that DBMS/Copy generates, are not guaranteed to be able to be opened by SAS itself. Hence our need to install SAS here. Theoretically, the .sas7bdat files are supposed to be cross-platform compatible and I think Peter had commented on that in a prior thread on this subject. However, hands on experience has suggested that this expectation is not absolute. HTH, Marc Schwartz I have had good success with Stat/Transfer (and its Windows version runs perfectly under Linux using wine) but haven't tested it as extensively as you've tested DBMS/Copy. Frank -- Frank E Harrell Jr Professor and Chair School of Medicine Department of Biostatistics Vanderbilt University __ 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] Pause loop
On Fri, 2007-12-28 at 19:07 +0100, Armin Goralczyk wrote:
Hi list
How can I pause a loop (or any other function) for a defined time? E.g.
for(i in 1:5) {print(1:i); 'function to pause for 10 seconds'}
Thanks for help.
At least on Linux, see:
?Sys.sleep
Note the caveats regarding time resolution...
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.
[R] unit attribute to list elements
Hi, I've started my own (first) package, part of which consists in listing common physical constants (Planck's constant, the speed of light in vacuum, etc). I'm wondering what would be a good way of dealing with pairs of value/unit. constants - list( cel = 2.99792458e8 , #m/s Z0 = 376.730313461, #ohm eps0 = 8.854187817e-12,#F/m mu0 = 4*pi*1e-7,#N/A^2 G = 6.67428e-11 # m^3 kg-1 s-2 ) I thought I could include the unit in the names attribute of each element, as in : names(constants$cel)- speed of light in vacuum [m.s^-1] Writing this for every element is very redundant... Is there any way to access and set the name of each first level element of the list? namesFirstLevelElements(constants)- c( speed of light in vacuum [m.s^-1], impedance of vacuum [some unit], ...) Quite possibly, I'm completely on the wring track; - maybe such a package already exists - a custom class or structure would be more appropriate? I don't really know how to deal with classes, though, and I'd like to keep the access to the constants' values as direct and generic as possible. Many thanks in advance, baptiste _ Baptiste Auguié Physics Department University of Exeter Stocker Road, Exeter, Devon, EX4 4QL, UK Phone: +44 1392 264187 http://newton.ex.ac.uk/research/emag http://projects.ex.ac.uk/atto __ 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] logistic mixed effects models with lmer
On Dec 28, 2007 9:35 AM, Sharon Goldwater [EMAIL PROTECTED] wrote: I have a question about some strange results I get when using lmer to build a logistic mixed effects model. I have a data set of about 30k points, and I'm trying to do backwards selection to reduce the number of fixed effects in my model. I've got 3 crossed random effects and about 20 or so fixed effects. At a certain point, I get a model (m17) where the fixed effects are like this (full output is at end of message): print(m17, corr=F) ... Fixed effects: Estimate Std. Error z value Pr(|z|) (Intercept)-1.978870.19699 -10.045 2e-16 *** sexM0.455530.14387 3.166 0.001544 ** ... is_discTRUE 0.246760.15204 1.623 0.104576 poly(wfreq, 2)1 -119.72397 11.00516 -10.879 2e-16 *** poly(wfreq, 2)217.356465.44456 3.188 0.001433 ** poly(wlen_p, 2)1 -13.607987.26926 -1.872 0.061208 . poly(wlen_p, 2)2 -6.431675.24119 -1.227 0.219770 ... where poly(wlen_p,2)2 is the least significant factor left in the model. So I then build a model (m18) with exactly the same random and fixed effects except removing poly(wlen_p,2)2. Then I do an anova, and I get: anova(m17,m18) Data: Models: m18: is_err ~ sex + starts_turn + before_hes + after_hes + before_part + m17: after_part + first_rep + is_open + is_disc + poly(wfreq, m18: 2) + wlen_p + poly(utt_rate, 2) + poly(dur, 2) + pmean + m17: poly(log_prange, 2) + poly(imean, 2) + poly(irange, 2) + m18: (1 | speaker) + (1 | corpus) + (1 | ref) m17: is_err ~ sex + starts_turn + before_hes + after_hes + before_part + m18: after_part + first_rep + is_open + is_disc + poly(wfreq, m17: 2) + poly(wlen_p, 2) + poly(utt_rate, 2) + poly(dur, 2) + m18: pmean + poly(log_prange, 2) + poly(imean, 2) + poly(irange, m17: 2) + (1 | speaker) + (1 | corpus) + (1 | ref) DfAICBIC logLik Chisq Chi Df Pr(Chisq) m18 27 25928 26153 -12937 m17 28 25925 26159 -12934 5.2136 10.02241 * So my first question is: Should I be concerned that the significance level shown in the original m17 is so different from the one shown by the anova? It's hard for me to see how this could happen. I noticed that there is a post on the FAQ about significance levels in linear mixed models, but I'm not sure whether it applies to the logistic case and if so how. It can happen that the significance levels reported by the different tests will be different. The test in the summary is based on a local approximation to the conditional mean and assumes that the variances of the random effects will not change substantially when fitting the model with and without that term. Did they? One thing I noticed is your output (and thank you for including that) is that it reports that corpus has only two levels. Is that correct? If so, I would not advise using a random effect for corpus. It is very difficult to estimate a variance from only two levels of a factor. I suggest using a fixed effect for corpus instead. Now, my second question is a result of removing one more factor (is_disc) from the model, creating m19. I do another anova: anova(m19,m18) Data: Models: m19: is_err ~ sex + starts_turn + before_hes + after_hes + before_part + m18: after_part + first_rep + is_open + poly(wfreq, 2) + wlen_p + m19: poly(utt_rate, 2) + poly(dur, 2) + pmean + poly(log_prange, m18: 2) + poly(imean, 2) + poly(irange, 2) + (1 | speaker) + (1 | m19: corpus) + (1 | ref) m18: is_err ~ sex + starts_turn + before_hes + after_hes + before_part + m19: after_part + first_rep + is_open + is_disc + poly(wfreq, m18: 2) + wlen_p + poly(utt_rate, 2) + poly(dur, 2) + pmean + m19: poly(log_prange, 2) + poly(imean, 2) + poly(irange, 2) + m18: (1 | speaker) + (1 | corpus) + (1 | ref) DfAICBIC logLik Chisq Chi Df Pr(Chisq) m19 26 25925 26142 -12936 m18 27 25928 26153 -12937 0 1 1 and now it seems that m19 (which contains fewer parameters than m18) is a better fit. I don't see how it's possible to remove parameters from a model and get a better likelihood, but I will confess that I don't entirely understand how these kinds of models are estimated. Does this have something to do with approximations that R is making to fit the models, or numerical rounding errors? Could either problem be due to correlations among variables? All my fixed effects are either binary or numeric, and there are some fairly high correlations between a few pairs of them (maybe as high as .6 or .65 using Kendall tau) but I figured that this would be ok given the large number of data points. I think each value of each binary feature is observed at least 70-100 times. The difference could be due to approximations or due to poor convergence. For some models the Laplace approximation can be generalized to a
[R] SAS to R - if you have SAS 8.2+
Hi there,
the attached R function uses the SAS Integrated Object
Model (IOM) and it can deal with SAS dates and long
variable names. All you need to provide is the folder
where the SAS data file is and the data file name
without the extension. The function requires the rcom
package.
This is meant to be first cut...but improvements and
suggestions are more than welcome!
Gyula
import.sas.data -
function(inPath,inSAS,as.is=FALSE){
# Function to read in a SAS data file
# Packages needed: rcom
# inPath: path to SAS file
# inSAS: SAS data file name (no extension)
# as.is: as per read.csv (FALSE - force all character
variables to be factors)
# outputs a data.frame object
# Created: December 1, 2007
# Modified: December 4, 2007
# Author: Gyula Gulyas ([EMAIL PROTECTED])
# Use as is, no guarantees, liability due to data
loss is limited to the price you paid for this
function...
require(rcom)
# check if user has closing slash in path and fix it
if needed
inPath - sub(/$,,inPath)
obWSM -
comCreateObject(SASWorkspaceManager.WorkspaceManager)
obWSM.Workspaces - obWSM[[Workspaces]]
obSAS - comCreateObject(SAS.Workspace)
obSAS.DataService - obSAS[[DataService]]
obSAS.LanguageService - obSAS[[LanguageService]]
# hard-coded temporary files
# sas temporary csv file
csvdata - paste(inPath,/t__sd__t.csv, sep=)
# sas temporary column definition file
coldef - paste(inPath,/t__sc__t.csv, sep=)
libRef -
obSAS.DataService$AssignLibref(sasds,,inPath,)
# create the content csv file
cont1 - paste(proc contents data=sasds.,inSAS,
out=_tmp1(KEEP=NAME TYPE LENGTH VARNUM FORMAT)
noprint; run;,sep=)
cont2 - proc sort data=_tmp1; by varnum; run;
cont3 - data _tmp2; set _tmp1; if type=2 then
dummy='character';
cont3 - paste(cont3,if type=1 then do; if format in
('DATE','DATETIME') then dummy='date'; ,sep=)
cont3 - paste(cont3,else dummy='numeric'; end;
run;,sep=)
cont4 - proc transpose data=_tmp2
out=_tmp3(DROP=_NAME_); id name; var dummy; run;
cont5 - paste(proc export data=_tmp3
outfile=',coldef,' dbms=csv; run;,sep=)
obSAS.LanguageService$Submit(cont1)
obSAS.LanguageService$Submit(cont2)
obSAS.LanguageService$Submit(cont3)
obSAS.LanguageService$Submit(cont4)
obSAS.LanguageService$Submit(cont5)
# column definitions, all as characters
scf - read.csv(coldef,as.is=T)
sasline1 - paste(data _tmp4; set sasds., inSAS, ;
run;,sep=)
obSAS.LanguageService$Submit(sasline1)
sasline2 - paste(proc export data=_tmp4
outfile=',csvdata,' dbms=csv; run;,sep=)
obSAS.LanguageService$Submit(sasline2)
sdf - read.csv(csvdata,as.is=as.is)
# delete old csvdata file
file.remove(csvdata)
# delete old coldef file
file.remove(coldef)
# convert SAS DATETIME format into native R date
(POSIXct)
x - which(scf==date)
sdf[,x] - as.data.frame(strptime(sdf[,x],
%d%b%Y:%H:%M:%S))
# cleanup SAS objects
obSAS.UniqueIdentifier - obSAS[[UniqueIdentifier]]
obWSM.Workspaces$RemoveWorkspaceByUUID(obSAS.UniqueIdentifier)
obSAS$Close()
return(sdf)
}
# example call
# not run
# test- import.sas.data(path/to/data,sasdatafile)
Looking for last minute shopping deals?
__
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] AIC and BIC in mixed effects model
Dear R Users: I am trying to compare several structures of the within-patient covariance such as unstructured, Autoregressive, and spatial by using the MIXED effects model. Can AIC, BIC be negative ? If yes, then in what situations they may be negative. Thanks a lot. [[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] Return Value of TCl/Tk window in R
res - tkmessageBox(title = test,message = Continue?, +icon =question, type = okcancel) if (tclvalue(res) == ok) 1 else 2 Happy new year! Philippe Grosjean Richard Müller wrote: Hello, I have the TCl/Tk command tkmessageBox(titel=,message=x,icon=question,type=okcancel) in my R script. Now I want to perform some operation in relation to the user's choice, something like if (okpressed) xxx else yyy What values does this command give and how are they used? Thank you, Richard __ 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] gplot function - error
Hi all, i installed the ggplot2 package via install.packages(), but the gplot function was not recognized in R console command. Is there any paths to configure? The error message reports that the function was not found. Thanks Ricardo Be a better friend, newshound, and __ 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] Conditionally incrementing a loop counter: Take 2
Dear Mike,
You could use a repeat loop and manage the index yourself:
i - 0
repeat{
x - runif(1)
if (x .1){
i - i + 1
cat(x = , x, \n)
}
if (i == 10) break
}
But if your example problem reflects your actual application, why not just
generate uniform random numbers on the interval (0, .1)?
I hope this helps,
John
John Fox, Professor
Department of Sociology
McMaster University
Hamilton, Ontario, Canada L8S 4M4
905-525-9140x23604
http://socserv.mcmaster.ca/jfox
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
project.org] On Behalf Of Mike Jones
Sent: December-27-07 6:08 PM
To: Peter Dalgaard
Cc: [EMAIL PROTECTED]
Subject: Re: [R] Conditionally incrementing a loop counter: Take 2
Since I didn't want the i to increment in the loop when the condition
is not met, then in my example I wanted the loop to actually run 14
times instead of the 10 since I wanted 4 of the iterations to be thrown
away, or ignored. I still haven't been able to figure this out. Going
the while route doesn't seem to work for me either.
nums - numeric(10)
i - 1
garbage - 0
while (i = 10){
x - runif(1)
cat(x = ,x,\n)
if (x 0.1){
nums[i] - x
i - i + 1
}
else{
garbage - garbage+1
}
cat(i = ,i,garbage = ,garbage,\n)
}
-Original Message-
From: Peter Dalgaard [mailto:[EMAIL PROTECTED]
Sent: Thursday, December 27, 2007 5:36 PM
To: Mike Jones
Cc: [EMAIL PROTECTED]
Subject: Re: [R] Conditionally incrementing a loop counter: Take 2
Mike Jones wrote:
My apologies for not including a working example.
Here it is:
for (i in 1:10){
cat(initial i = ,i,\n)
x - runif(1)
if (x 0.7){
i - i-1
}
cat(second i = ,i,\n)
}
When I ran this i got what follows, so there were four cases where I
wanted the i not to increment.
initial i = 1
second i = 1
initial i = 2
second i = 1
initial i = 3
second i = 3
initial i = 4
second i = 3
initial i = 5
second i = 4
initial i = 6
second i = 6
initial i = 7
second i = 7
initial i = 8
second i = 7
initial i = 9
second i = 9
initial i = 10
second i = 10
Is this the kind of effect you want?
x - runif(10)
cbind(x, 1:10, cumsum(x .7))
x
[1,] 0.384165631 1 1
[2,] 0.392715845 2 2
[3,] 0.895936431 3 2
[4,] 0.910242185 4 2
[5,] 0.689987301 5 3
[6,] 0.237071326 6 4
[7,] 0.225032680 7 5
[8,] 0.001856286 8 6
[9,] 0.392034868 9 7
[10,] 0.655076045 10 8
If you insist on using a loop, you need to separate the loop control
from the manipulation of i, as in (e.g.)
i - 0
for (j in 1:10){
i - i + 1
cat(initial i = ,i,\n)
x - runif(1)
if (x 0.7){
i - i-1
}
cat(second i = ,i,\n)
}
-Original Message-
From: Mike Jones
Sent: Thursday, December 27, 2007 4:35 PM
To:'[EMAIL PROTECTED]'
Subject: Conditionally incrementing a loop counter
Hi,
I am trying a for loop from 1 to 10 by 1. However, if a condition
does not get met, I want to throw away that iteration. So if my
loop is for (i in 1:10) and i is say, 4 and the condition is not met
then I don't want i to go up to 5. Is there a way to do that? I
can't seem to manually adjust i because from what I understand, R
creates 10 long vector and uses that to loops thru and I'm not
sure
how to get at the index of that vector. Any suggestions? Thanks in
advance.
Mike Jones
Westat
1650 Research Blvd. RE401
Rockville, MD 20850
Ph: 240.314.2312
[[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.
--
O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B
c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K
(*) \(*) -- University of Copenhagen Denmark Ph: (+45)
35327918
~~ - ([EMAIL PROTECTED]) FAX: (+45)
35327907
__
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] groupedData function not found
Hi Andrea, did you try require(nlme) ? Adding a package is not necessarily the same as loading it. Cheers Andrew On Thu, Dec 27, 2007 at 10:42:27PM +, andrea previtali wrote: Hello, I'm trying to use the groupedData function and R is giving me the message: Error: can not find function groupedData The code is coming from a textbook so I think it should be correct: pigs-data.frame(cbind(pig.time,pig.id,pig.wt)) pig.growth-groupedData(pig.wt~pig.time|pig.id,data=pigs) I have added the package nlme and to confirm that it was installed correctly I requested the list of functions included in the package (library(help=nlme)) and I do see groupData in the list. I am using R 2.6.1 in Windows XP. I'll appreciate your help. Thanks, Andrea Previtali Post-doc fellow Dept. of Biology, Univ. of Utah, SLC, UT. __ 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. -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. -- Andrew Robinson Department of Mathematics and StatisticsTel: +61-3-8344-9763 University of Melbourne, VIC 3010 Australia Fax: +61-3-8344-4599 http://www.ms.unimelb.edu.au/~andrewpr http://blogs.mbs.edu/fishing-in-the-bay/ __ 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] unit attribute to list elements
Is this what you want? Lines - cel 3.0 m/s + Z0 367 ohm + eps0 8.9e-12 F/m + Constants.DF - read.table(textConnection(Lines), as.is = TRUE) Constants - as.list(Constants.DF[[2]]) names(Constants) - Constants.DF[[1]] for(i in seq_along(Constants)) comment(Constants[[i]]) - Constants.DF[i, 3] # get value Constants$Z0 [1] 367 # get comment comment(Constants$Z0) [1] ohm # add another Constant Constants$e - 2.7 comment(Constants$e) - exp On Dec 28, 2007 2:33 PM, baptiste Auguié [EMAIL PROTECTED] wrote: Hi, I've started my own (first) package, part of which consists in listing common physical constants (Planck's constant, the speed of light in vacuum, etc). I'm wondering what would be a good way of dealing with pairs of value/unit. constants - list( cel = 2.99792458e8 , #m/s Z0 = 376.730313461, #ohm eps0 = 8.854187817e-12,#F/m mu0 = 4*pi*1e-7,#N/A^2 G = 6.67428e-11 # m^3 kg-1 s-2 ) I thought I could include the unit in the names attribute of each element, as in : names(constants$cel)- speed of light in vacuum [m.s^-1] Writing this for every element is very redundant... Is there any way to access and set the name of each first level element of the list? namesFirstLevelElements(constants)- c( speed of light in vacuum [m.s^-1], impedance of vacuum [some unit], ...) Quite possibly, I'm completely on the wring track; - maybe such a package already exists - a custom class or structure would be more appropriate? I don't really know how to deal with classes, though, and I'd like to keep the access to the constants' values as direct and generic as possible. Many thanks in advance, baptiste _ Baptiste Auguié Physics Department University of Exeter Stocker Road, Exeter, Devon, EX4 4QL, UK Phone: +44 1392 264187 http://newton.ex.ac.uk/research/emag http://projects.ex.ac.uk/atto __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] unit attribute to list elements
On 28 Dec 2007, at 20:06, Gabor Grothendieck wrote: Is this what you want? Perfect! Thanks a lot! Lines - cel 3.0 m/s + Z0 367 ohm + eps0 8.9e-12 F/m + Constants.DF - read.table(textConnection(Lines), as.is = TRUE) Constants - as.list(Constants.DF[[2]]) names(Constants) - Constants.DF[[1]] for(i in seq_along(Constants)) comment(Constants[[i]]) - Constants.DF[i, 3] # get value Constants$Z0 [1] 367 # get comment comment(Constants$Z0) [1] ohm # add another Constant Constants$e - 2.7 comment(Constants$e) - exp On Dec 28, 2007 2:33 PM, baptiste Auguié [EMAIL PROTECTED] wrote: Hi, I've started my own (first) package, part of which consists in listing common physical constants (Planck's constant, the speed of light in vacuum, etc). I'm wondering what would be a good way of dealing with pairs of value/unit. constants - list( cel = 2.99792458e8 , #m/s Z0 = 376.730313461, #ohm eps0 = 8.854187817e-12,#F/m mu0 = 4*pi*1e-7,#N/A^2 G = 6.67428e-11 # m^3 kg-1 s-2 ) I thought I could include the unit in the names attribute of each element, as in : names(constants$cel)- speed of light in vacuum [m.s^-1] Writing this for every element is very redundant... Is there any way to access and set the name of each first level element of the list? namesFirstLevelElements(constants)- c( speed of light in vacuum [m.s^-1], impedance of vacuum [some unit], ...) Quite possibly, I'm completely on the wring track; - maybe such a package already exists - a custom class or structure would be more appropriate? I don't really know how to deal with classes, though, and I'd like to keep the access to the constants' values as direct and generic as possible. Many thanks in advance, baptiste _ Baptiste Auguié Physics Department University of Exeter Stocker Road, Exeter, Devon, EX4 4QL, UK Phone: +44 1392 264187 http://newton.ex.ac.uk/research/emag http://projects.ex.ac.uk/atto __ 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] gplot function - error
You changed the intial letter: the function is qplot and not gplot. domenico Ricardo Perrone wrote: Hi all, i installed the ggplot2 package via install.packages(), but the gplot function was not recognized in R console command. Is there any paths to configure? The error message reports that the function was not found. Thanks Ricardo Be a better friend, newshound, and __ 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] Forcing virtual digits
Hi there, I'm using XP with R 2.6.1 I've got the following question: Is there a way to force virtual digits? I mean: In the external csv-file, I read in some numbers, e.g. 33 which is written into a variable. I want to add this variable into a plot with an additional digit: 33.0. I know, that this is a kind of cheating, because the number was not measured as precisely, but it looks better in the graph. Any ideas? Thanks for your help, Thomas [[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] two plots on the same page
I'd like to know why I cannot get a plot and the QQnorm in the same sheet. The commands are simple but: library(nlme) glmod1 - gls(upfmla,correlation=corAR1(),method=ML) summary(glmod1) par(mfrow = c(2,1)) plot(glmod1, main=GLS Residuals vs. GLS Fitted) qqnorm(glmod1) No matter what (I tried different permutations of the plotting commands) the second drawing is overlapped to the first one instead of being placed underneath it on the same page. How come? Thank you very much. Happy New Year, Maura -- Maura E.M [[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] print ggplot2 summary on graphic
Hi all: The code below works well to print annotations on base graphics but I can't make it work using the ggplot2 package. I tried to use the capture.output function but only prints the summary on the R console and no on the graphic. Any ideas on how to print output on ggplot2 graphs? For a powerpoint presentation I basically just want to show viewers the mean and Median on the graph since they won't be seeing the data or the R console. aa - 1:10 plot(aa) text(c(2,4),10, labels=c( mean, sd),col=2) text(c(2,4),9.5, labels=c(mean(aa), sd(aa)),col=4) legend(topright,capture.output(sd(aa[1:2]))) m - ggplot(movies, aes(x=rating)) m + geom_density() capture.output(mean(movies[3])) Felipe D. Carrillo Fishery Biologist US Fish Wildlife Service California, USA Looking for last minute shopping deals? __ 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] FYI: Package installation problem with windows
Hello, maybe this info is helpful for someone else, at least it is a reminder for me if the problem should reoccur: The last hours R drove me nuts because it wouldn't install new packages, or more irritating still: It would chose randomly which packages to install when given a list (unable to move temporary installation ). The solution was to turn of the indexation of folders. HTH Owe -- Owe Jessen Diplom-Volkswirt Hanssenstraße 17 24106 Kiel [EMAIL PROTECTED] http://www.econinfo.de __ 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] AIC and BIC in mixed effects model
Chuanjun Zhang zhangchu at umkc.edu writes: Dear R Users: I am trying to compare several structures of the within-patient covariance such as unstructured, Autoregressive, and spatial by using the MIXED effects model. Can AIC, BIC be negative ? If yes, then in what situations they may be negative. This almost deserves to be a FAQ, although it's a statistical rather than an R issue: http://finzi.psych.upenn.edu/R/Rhelp02a/archive/46734.html http://finzi.psych.upenn.edu/R/Rhelp02a/archive/108660.html The one-sentence answer is that since probability *densities* can be 1, log-likelihood *densities* can be 0 and hence negative log-likelihoods can be 0 (so AIC/BIC can also be 0). However, there's a potentially larger issue with using AICs for mixed models, which is that it's not always entirely clear what the right number of degrees of freedom is for a random effect ... cheers Ben Bolker __ 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 can I change the language back to english?
I have reset my windows and re-installed R (I explicitly asked for english) and although the windows XP version is in english I am having R display in french. like: R est un logiciel libre livré sans AUCUNE GARANTIE. Vous pouvez le redistribuer sous certaines conditions. Tapez 'license()' ou 'licence()' pour plus de détails. R est un projet collaboratif avec de nombreux contributeurs. Tapez 'contributors()' pour plus d'information et 'citation()' pour la façon de le citer dans les publications. Tapez 'demo()' pour des démonstrations, 'help()' pour l'aide en ligne ou 'help.start()' pour obtenir l'aide au format HTML. Tapez 'q()' pour quitter R. I know I have an azerty french keyboard but even before the reset I had the same azerty french keyborad and the same XP settings and R was completely in english, now it's annoyinly in french and reinstalling R again doesn't change it back to english like I want it. how can I set it in english, I hate having my R message in french. thanks. -- View this message in context: http://www.nabble.com/How-can-I-change-the-language-back-to-english--tp14532778p14532778.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] two plots on the same page
-Original Message- From: [EMAIL PROTECTED] on behalf of Maura E Monville Sent: Fri 12/28/2007 2:18 PM To: [EMAIL PROTECTED] Subject: [R] two plots on the same page I'd like to know why I cannot get a plot and the QQnorm in the same sheet. The commands are simple but: library(nlme) glmod1 - gls(upfmla,correlation=corAR1(),method=ML) summary(glmod1) par(mfrow = c(2,1)) plot(glmod1, main=GLS Residuals vs. GLS Fitted) qqnorm(glmod1) No matter what (I tried different permutations of the plotting commands) the second drawing is overlapped to the first one instead of being placed underneath it on the same page. How come? Because glmod1 is of class gls class(glmod1) [1] gls so when you plot it you are invoking plot.gls() See the help for function plot.gls() in the nlme library. ?plot.gls reports that the value of a call is Value a diagnostic Trellis plot. so you are dealing with Trellis plots handled by the lattice package. Trellis plots do not use the par() settings used by regular plots. Here's one way to do it. (I substituted an example from nlme since I don't have your data and upfmla object. library(nlme) glmod1 - gls(follicles ~ sin(2*pi*Time) + cos(2*pi*Time), Ovary, correlation = corAR1(form = ~ 1 | Mare)) summary(glmod1) plot1 - plot(glmod1, main=GLS Residuals vs. GLS Fitted) plot2 - qqnorm(glmod1) print(plot1, split = c(1, 1, 1, 2), more = TRUE) print(plot2, split = c(1, 2, 1, 2)) Try something like this. The 'split' mechanism works for me and I get both plots on one page. Remember also that you can save the output of trellis plot commands and print them later. Trellis plots do not always print right away - in many situations you must explicitly wrap your trellis plot command with print() before it will render on a graphics device. You can read more about putting multiple trellis plots on a page in the help page for print.trellis in the lattice package. Hope this helps. After you master the lattice package you can go to that New Years party and celebrate! Steven McKinney Statistician Molecular Oncology and Breast Cancer Program British Columbia Cancer Research Centre email: smckinney +at+ bccrc +dot+ ca tel: 604-675-8000 x7561 BCCRC Molecular Oncology 675 West 10th Ave, Floor 4 Vancouver B.C. V5Z 1L3 Canada Thank you very much. Happy New Year, Maura -- Maura E.M [[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] Forcing virtual digits
?sprintf x - 33 plot(0, main=sprintf(Value = %.1f, x)) On Dec 28, 2007 5:13 PM, Thomas Schwander [EMAIL PROTECTED] wrote: Hi there, I'm using XP with R 2.6.1 I've got the following question: Is there a way to force virtual digits? I mean: In the external csv-file, I read in some numbers, e.g. 33 which is written into a variable. I want to add this variable into a plot with an additional digit: 33.0. I know, that this is a kind of cheating, because the number was not measured as precisely, but it looks better in the graph. Any ideas? Thanks for your help, Thomas [[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 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] FYI: Package installation problem with windows
On Fri, 28 Dec 2007, Owe Jessen wrote: Hello, maybe this info is helpful for someone else, at least it is a reminder for me if the problem should reoccur: The last hours R drove me nuts because it wouldn't install new packages, or more irritating still: It would chose randomly which packages to install when given a list (unable to move temporary installation ). The solution was to turn of the indexation of folders. See the rw-FAQ Q4.8. It is not normal for the Windows' Indexing Service to cause any problems here, but enhanced versions can. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ 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 can I change the language back to english?
This is rw-FAQ Q3.2, Q3.4. But the underlying problem is that Windows is telling R it is in French, and perhaps you need to sort that out. On Fri, 28 Dec 2007, francogrex wrote: I have reset my windows and re-installed R (I explicitly asked for english) and although the windows XP version is in english I am having R display in french. like: R est un logiciel libre livré sans AUCUNE GARANTIE. Vous pouvez le redistribuer sous certaines conditions. Tapez 'license()' ou 'licence()' pour plus de détails. R est un projet collaboratif avec de nombreux contributeurs. Tapez 'contributors()' pour plus d'information et 'citation()' pour la façon de le citer dans les publications. Tapez 'demo()' pour des démonstrations, 'help()' pour l'aide en ligne ou 'help.start()' pour obtenir l'aide au format HTML. Tapez 'q()' pour quitter R. I know I have an azerty french keyboard but even before the reset I had the same azerty french keyborad and the same XP settings and R was completely in english, now it's annoyinly in french and reinstalling R again doesn't change it back to english like I want it. how can I set it in english, I hate having my R message in french. thanks. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595__ 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.
