R: [R] solving for a transition point in a piecewise nonlinear model
Dear Robert, In general it may be difficult to estimate a model with generic (possibly nonlinear) functions before/after the changepoint to be estimated too. However if you are willing to make some restrinctions on your F1(.) and F2(.), you could semplify the problem.. For instance, have a look to the strucchange and segmented packages. The former mainly deals with time series data, the latter with GLM. hope this helps, vito PS Of course you can always perform a grid-search algorithm by fitting different models for each fixed alpha, and picking the value that maximizes the Lik.. - Original Message - From: Robert Musk [EMAIL PROTECTED] To: [EMAIL PROTECTED] Sent: Wednesday, July 14, 2004 6:43 AM Subject: [R] solving for a transition point in a piecewise nonlinear model I want to fit a piecewise nonlinear model in which the transition point is an estimated parameter. Something like: F(x)=F1(x) when xalpha, F(x)=F2(x) when x=alpha. How can I solve for alpha within the nls call? Thanks, Rob __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] summary() doesn't work with Date class objects
Scott Waichler wrote:
The handy function summary() doesn't work correctly with Date class
objects:
R.version.string
[1] R version 1.9.1, 2004-06-21
b - as.Date(c(2002-12-26, 2002-12-27, 2002-12-28, 2002-12-29, 2002-12-30))
b
[1] 2002-12-26 2002-12-27 2002-12-28 2002-12-29 2002-12-30
summary(b)
Min. 1st Qu. Median Mean 3rd Qu. Max.
2002-12-29 2002-12-29 2002-12-29 2002-12-29 2002-12-29 2002-12-29
The obvious fix is to change summary.date (in
.../src/library/base/R/dates.R) as follows:
old:
summary.Date - function(object, ...)
{
x - summary.default(unclass(object), ...)[1:6]# not NA's
class(x) - oldClass(object)
x
}
new:
summary.Date - function(object, ...)
{
x - unclass(object)
x - summary.default(x, digits = floor(log(x)) + 1, ...)[1:6]# not NA's
class(x) - oldClass(object)
x
}
One might want to change floor(log(x)) + 1 to something less
computational in intensive like 10. ;-)
Uwe Ligges
BTW: Should I submit a bug report or does anybody fix the sources at once?
Scott Waichler
Pacific Northwest National Laboratory
Richland, WA 99352USA
[EMAIL PROTECTED]
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] RGui Titlebar
In the windows version (RGui), is there a way to set the text displayed in the titlebar of the R window? When I have 2 instances of RGui running, it would be helpul if the titlebar could help to understand which is which. -- Erich Neuwirth, Computer Supported Didactics Working Group Visit our SunSITE at http://sunsite.univie.ac.at Phone: +43-1-4277-38624 Fax: +43-1-4277-9386 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] A function likes table
DeaR useRs: Excuse me for my bad English. I am looking for a function likes table or ftable which returns the sum in a object by the cross tabulation, not the counts of the cross tabulation. I dont know if everybody understand me. Thank you very much. Agus --- [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Re:
use unique(data) or table(data) __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Permutations
Jordi:
If I understand you well, the function below may do what you asked for.
It is not clear to me from your posting wether e.g.
1 2 4 3 5 6 7 8 910 11 12
and
1 4 2 3 5 6 7 8 910 11 12
should count as differente permutations, i.e., wether once one pair of
elements interchange their
blocks, permutations within any block are allowable. I have assumed that
only one pair of
elements are interchanged (but the function could be modified to account
for other possibilities).
permutations
function(elements,blocks) {
n - length(elements)
el.per.block - n / blocks
for (i in 1:n) {# For each element in turn,
b - floor(i/(el.per.block+.1))+1 # find which block it belongs to.
if (b==blocks)# If in the last block, we are done.
break
allow.pos - b*el.per.block + 1 # Find first position it could
migrate to...
for (j in (allow.pos:n)) {# and create permutations with
all allowable
perm - elements# interchanges.
perm[i] - elements[j]
perm[j] - elements[i]
print(perm)
}
}
}
permutations(1:4,2)
[1] 3 2 1 4
[1] 4 2 3 1
[1] 1 3 2 4
[1] 1 4 3 2
permutations(1:6,2)
[1] 4 2 3 1 5 6
[1] 5 2 3 4 1 6
[1] 6 2 3 4 5 1
[1] 1 4 3 2 5 6
[1] 1 5 3 4 2 6
[1] 1 6 3 4 5 2
[1] 1 2 4 3 5 6
[1] 1 2 5 4 3 6
[1] 1 2 6 4 5 3
permutations(1:9,3)
[1] 4 2 3 1 5 6 7 8 9
[1] 5 2 3 4 1 6 7 8 9
[1] 6 2 3 4 5 1 7 8 9
[1] 7 2 3 4 5 6 1 8 9
[1] 8 2 3 4 5 6 7 1 9
[1] 9 2 3 4 5 6 7 8 1
[1] 1 4 3 2 5 6 7 8 9
[1] 1 5 3 4 2 6 7 8 9
[1] 1 6 3 4 5 2 7 8 9
[1] 1 7 3 4 5 6 2 8 9
[1] 1 8 3 4 5 6 7 2 9
[1] 1 9 3 4 5 6 7 8 2
[1] 1 2 4 3 5 6 7 8 9
[1] 1 2 5 4 3 6 7 8 9
[1] 1 2 6 4 5 3 7 8 9
[1] 1 2 7 4 5 6 3 8 9
[1] 1 2 8 4 5 6 7 3 9
[1] 1 2 9 4 5 6 7 8 3
[1] 1 2 3 7 5 6 4 8 9
[1] 1 2 3 8 5 6 7 4 9
[1] 1 2 3 9 5 6 7 8 4
[1] 1 2 3 4 7 6 5 8 9
[1] 1 2 3 4 8 6 7 5 9
[1] 1 2 3 4 9 6 7 8 5
[1] 1 2 3 4 5 7 6 8 9
[1] 1 2 3 4 5 8 7 6 9
[1] 1 2 3 4 5 9 7 8 6
Notice that no error checking of any kind is done: one should check,
e.g. that el.per.block is
integer.
Best,
ft.
--
Fernando TUSELLe-mail:
Departamento de Econometría y Estadística [EMAIL PROTECTED]
Facultad de CC.EE. y Empresariales Tel: (+34)94.601.3733
Avenida Lendakari Aguirre, 83 Fax: (+34)94.601.3754
E-48015 BILBAO (Spain)Secr: (+34)94.601.3740
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Permutations
Perhaps what you want might better be described as ordered partitions? Is what you want the following: We study sequences of length 12 and divide them in 4 segments position 1 2 3, position 4 5 6, position 7 8 9, position 10 11 12, Find all permutation sequences of the numbers 1 to 12 with the property that all segment sequences are monotonically increasing. I think that produces what you need. Since the segments are ordered, you avoid intra-block permutations. If that is what you want, writing a recursive function should not be too hard. Jordi Altirriba Gutirrez wrote: Dear R users, Im a beginner user of R and Ive a problem with permutations that I dont know how to solve. Ive 12 elements in blocks of 3 elements and I want only to make permutations inter-blocks (no intra-blocks) (sorry if the terminology is not accurate), something similar to: 1 2 3 | 4 5 6 | 7 8 9 | 10 11 12 --1st permutation 1 3 2 | 4 5 6 | 7 8 9 | 10 11 12 NO - - 3 2 1 | 4 5 6 | 7 8 9 | 10 11 12 NO - - - 1 2 4 | 3 5 6 | 7 8 9 | 10 11 12 YES-2nd permutation -- 4 5 6 | 1 2 3 | 7 8 9 | 10 11 12 YES-3rd permutation - - - - - - 4 5 6 | 2 1 3 | 7 8 9 | 10 11 12 NO - - Thanks for your time, Jordi Altirriba Ph D student Hospital Clinic Barcelona - Spain MSN Motor. http://motor.msn.es/researchcentre/ __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Erich Neuwirth, Computer Supported Didactics Working Group Visit our SunSITE at http://sunsite.univie.ac.at Phone: +43-1-4277-38624 Fax: +43-1-4277-9386 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] RGui Titlebar
It look like this will be possible from R v2.0.0; From http://cran.r-project.org/bin/windows/base/CHANGES.rw2000dev Added functions setWindowTitle(), getWindowTitle(), and getIdentification(). To R-devel: Will this be 1) Rgui only? and/or 2) Windows only? Cheers Henrik Bengtsson -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Erich Neuwirth Sent: Wednesday, July 14, 2004 8:57 AM To: [EMAIL PROTECTED] Subject: [R] RGui Titlebar In the windows version (RGui), is there a way to set the text displayed in the titlebar of the R window? When I have 2 instances of RGui running, it would be helpul if the titlebar could help to understand which is which. -- Erich Neuwirth, Computer Supported Didactics Working Group Visit our SunSITE at http://sunsite.univie.ac.at Phone: +43-1-4277-38624 Fax: +43-1-4277-9386 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] SWIG for R
Hi. Has R dev. team considered employing SWIG(http://www.swig.org/), which supports PHP, Ruby, Java etc., for connecting C/C++ libraries with R? Regards. __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] (no subject) (was: Permutations)
Jordi try this R x - c(1,2,3, 10,11,12, 41,42,43, 81,82,83) R dim(x) - c(3,4) R x [,1] [,2] [,3] [,4] [1,]1 10 41 81 [2,]2 11 42 82 [3,]3 12 43 83 R jj - t(apply(x,1,sample)) R jj [,1] [,2] [,3] [,4] [1,]1 41 10 81 [2,]2 11 82 42 [3,] 123 43 83 R as.vector(jj) R [1] 1 2 12 41 11 3 10 82 43 81 42 83 and I think that does what you want... We take the vector, rearrange it into a matrix with three rows, then sample *within* the rows, then rearrange into a vector again. There will be one forbidden permutation, namely the identity (which may or may not be desirable). This method doesn't allow intra block permutations. best rksh Dear R users, First of all, thanks for the incredibly fast answers and help of Rolf, Marc and Robert. Yes, I noticed that it was a lot of permutacions, but my intention was to make this process automatic and take only 5.000 - 10.000 permutations. Therefore, I wanted only to take that interesting permutations with some information [inter-block permutations]. The reason why I'm interested in these permutations is because I'm using some packages of Bioconductor to analyse my data from some microarrays and I thought that perhaps could be interesting to see what happens when I permute my data and I compare it against the not permuted data. Thanks again for your time and suggestions. Jordi Altirriba Ph. D. Student Hospital Clinic-Barcelona-Spain __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Robin Hankin Uncertainty Analyst Southampton Oceanography Centre SO14 3ZH tel +44(0)23-8059-7743 [EMAIL PROTECTED] (edit in obvious way; spam precaution) __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] ROracle - fetch gives empty dataframe
Dear R-users, I was able to make ROracle package to connect to the DB (Oracle91, 64bit, on Solaris). But after executing siple SQL query, fetch commaned gives me an empty dataframe. ### RORACLE INSTALATION PROCEDURE ### R CMD INSTALL --configure-args='--enable-extralibs' --enable-oracle32=no ~/tmp/ROracle_0.5-5.tar.gz #since we have 64bit Oracle9i instalation PROBLEM: after executing simple statement, from within R, fetch function gives me the empty dataframe: ### CONNECTION, and SQL QUERY EXECUTION ora - dbDriver(Oracle) con - dbConnect(ora, rado/only2admin dbListTables(con) character(0) rs - dbSendQuery(con, desc * from si_r where id=498) d - fetch(rs, n= -1) dim(d) [1] 0 3 str(d) data.frame': 0 obs. of 3 variables: $ ID : int $ DAY: chr $ R : num seems like connections has been established dbGetInfo(ora) $drvName [1] Oracle (ProC/C++) $connectionIds $connectionIds[[1]] OraConnection:(14939,0) $fetch_default_rec [1] 500 $managerId OraDriver:(14939) $length [1] 10 $num_con [1] 1 $counter [1] 1 $clientVersion [1] 0.5-4 Maybe it is something stupid (and I need just a hint), or may be it is something major. Thanks in advance, Rado Bonk -- Dr. Radoslav Bonk European Commission - DG Joint Research Centre (JRC) Institute for Environment and Sustainability (IES) LM Unit - Natural Hazards Weather Driven Natural Hazards Action Via E. Fermi, TP 261, 21020 Ispra (Va), Italy Tel.: 0039-0332-786013 Fax: 0039-0332-786653 Webpage: http://natural-hazards.jrc.it/floods/ __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] list of S3-methods
how can I get a list of all S3-methods (of a package) such that I know which functions to include in the S3method() in the NAMESPACE-file? Maybe separated by generic=T/F. thx Meinhard Ploner Vienna __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] RGui Titlebar
Henrik Bengtsson wrote: It look like this will be possible from R v2.0.0; From http://cran.r-project.org/bin/windows/base/CHANGES.rw2000dev Added functions setWindowTitle(), getWindowTitle(), and getIdentification(). To R-devel: Will this be 1) Rgui only? and/or 2) Windows only? Well, the title of that page already tells us Windows-specific changes to R Uwe Ligges Cheers Henrik Bengtsson -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Erich Neuwirth Sent: Wednesday, July 14, 2004 8:57 AM To: [EMAIL PROTECTED] Subject: [R] RGui Titlebar In the windows version (RGui), is there a way to set the text displayed in the titlebar of the R window? When I have 2 instances of RGui running, it would be helpul if the titlebar could help to understand which is which. -- Erich Neuwirth, Computer Supported Didactics Working Group Visit our SunSITE at http://sunsite.univie.ac.at Phone: +43-1-4277-38624 Fax: +43-1-4277-9386 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] ROracle - fetch gives...(corrected SQL code)
Sorry for posting the second time, I corrected the SQL code: PROBLEM: after executing simple statement, from within R, fetch function gives me the empty dataframe: ### CONNECTION, and SQL QUERY EXECUTION ora - dbDriver(Oracle) con - dbConnect(ora, rado/only2admin dbListTables(con) character(0) rs - dbSendQuery(con, select * from si_r where id=498) ^^ CORRECTION d - fetch(rs, n= -1) dim(d) [1] 0 3 str(d) data.frame': 0 obs. of 3 variables: $ ID : int $ DAY: chr $ R : num seems like connection has been established dbGetInfo(ora) $drvName [1] Oracle (ProC/C++) $connectionIds $connectionIds[[1]] OraConnection:(14939,0) $fetch_default_rec [1] 500 $managerId OraDriver:(14939) $length [1] 10 $num_con [1] 1 $counter [1] 1 $clientVersion [1] 0.5-4 Maybe it is something stupid (and I need just a hint), or may be it is something major. Thanks in advance, Rado Bonk -- Dr. Radoslav Bonk European Commission - DG Joint Research Centre (JRC) Institute for Environment and Sustainability (IES) LM Unit - Natural Hazards Weather Driven Natural Hazards Action Via E. Fermi, TP 261, 21020 Ispra (Va), Italy Tel.: 0039-0332-786013 Fax: 0039-0332-786653 Webpage: http://natural-hazards.jrc.it/floods/ __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] list of S3-methods
Meinhard Ploner wrote: how can I get a list of all S3-methods (of a package) such that I know which functions to include in the S3method() in the NAMESPACE-file? Maybe separated by generic=T/F. thx Meinhard Ploner Vienna Since one you does not register S3 methods (except for the Namespace file), you cannot get such a list very easily. You might want to look for function names with a dot in it and then look whether the second part of that name corresponds to a class. Uwe Ligges __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] RGui Titlebar
On Wed, 14 Jul 2004 09:55:26 +0200, Henrik Bengtsson [EMAIL PROTECTED] wrote: It look like this will be possible from R v2.0.0; From http://cran.r-project.org/bin/windows/base/CHANGES.rw2000dev Added functions setWindowTitle(), getWindowTitle(), and getIdentification(). To R-devel: Will this be 1) Rgui only? and/or 2) Windows only? From the man page: This sets the title of the frame in MDI mode, the title of the console for 'RGui --sdi', and the title of the window from which it was launched for 'Rterm'. It has no effect in embedded uses of R. Duncan Murdoch __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] list of S3-methods
Meinhard Ploner wrote: how can I get a list of all S3-methods (of a package) such that I know which functions to include in the S3method() in the NAMESPACE-file? Maybe separated by generic=T/F. thx Meinhard Ploner Vienna Since one you does not register S3 methods (except for the Namespace file), you cannot get such a list very easily. You might want to look for function names with a dot in it and then look whether the second part of that name corresponds to a class. Uwe Ligges Thank you. Usually users register S3-methods prior to make libraries or is this possible only for S4 classes? Meinhard Ploner __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] RGui Titlebar
An ugly workaround for versions prior 2.0 would be creating a copy of Rgui.exe and change its icon (google for 'change icon'). Under xp I also can just deactivate the icon on the compatibility tab (disable visual themes). Best Jens Oehlschlägel __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] list of S3-methods
Meinhard Ploner wrote: Meinhard Ploner wrote: how can I get a list of all S3-methods (of a package) such that I know which functions to include in the S3method() in the NAMESPACE-file? Maybe separated by generic=T/F. thx Meinhard Ploner Vienna Since one you does not register S3 methods (except for the Namespace file), you cannot get such a list very easily. You might want to look for function names with a dot in it and then look whether the second part of that name corresponds to a class. Uwe Ligges Thank you. Usually users register S3-methods prior to make libraries or is this possible only for S4 classes? Meinhard Ploner Yes. S3 method dispatchj is solely based on the naming conventions. Uwe __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] list of S3-methods
UweL == Uwe Ligges [EMAIL PROTECTED] on Wed, 14 Jul 2004 11:25:20 +0200 writes: UweL Meinhard Ploner wrote: how can I get a list of all S3-methods (of a package) such that I know which functions to include in the S3method() in the NAMESPACE-file? Maybe separated by generic=T/F. thx Meinhard Ploner Vienna UweL Since one you does not register S3 methods (except for UweL the Namespace file), you cannot get such a list very UweL easily. You might want to look for function names UweL with a dot in it and then look whether the second part UweL of that name corresponds to a class. yes; and for this you can re-use some of the code the current methods() function has. It *is* doing a very related thing: Looking for all S3 methods belonging to a given class or to a given generic. Martin Maechler __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] ROracle - fetch gives empty dataframe
Dear Rado, I think you didn't get a proper db connection at all. You forgot to specify the database to connect to. On Wed, 14 Jul 2004, Rado Bonk wrote: ### CONNECTION, and SQL QUERY EXECUTION ora - dbDriver(Oracle) con - dbConnect(ora, rado/only2admin Which database??? dbListTables(con) character(0) dbListTables would display something, if you had connected. You should get this working before you proceed. Regards, Michael __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] constrOptim and function with additional parameters?
How can I use a function with some additional input parameters
in constrOptim? For example, something like
fr - function(x,a) { ## Rosenbrock Banana function
x1 - x[1]
x2 - x[2]
a * (x2 - x1 * x1)^2 + (1 - x1)^2
}
where the optimum is to be found w.r.t. x. Calling
optim(c(-1.2,1), fr, NULL, a=100) works as expected, but I fail
to provide the a=100 in the constrained case:
constrOptim(c(-1.2,0.9), fr, NULL, ui=rbind(c(-1,0),c(0,-1)),
ci=c(-1,-1),a=100)
Error in f(theta) : Argument a is missing, with no default
Is this a bug or is there a different solution that I miss here?
TIA, Marlene
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Smooth monotone estimation on R
Browsing over the FDA book, I can not find any discussion of monotone smoothing. Andy From: Kjetil Halvorsen Hola! Experimenting a little, package fda seems very much under development, and help pages are definitely not finished. It would certainly help to read the book functional data analysis by the author of fda, which is in our library. On the other hand package mgcv is more mature, so it would seem easier to use that. It is not totally automatic, but the example library(mgcv) example(mono.con) has everything necessary to get started. Kjetil Halvorsen Eliyahu-Oron wrote: Kjetil and Andy, Thanks for your helpful answers! The first two (mgcv and fda) seem to be in the direction I'm looking for. I downloaded them both. I'm running into a lot of implementation difficulties, though. I wonder if there's anyone who tried to do a monotone spline using either the 'mgcv' or the 'fda' packages, whom I could ask directly? Thanks again, Assaf -Original Message- From: Kjetil Halvorsen [mailto:[EMAIL PROTECTED] Sent: Monday, July 12, 2004 4:32 PM To: Assaf P Oron Cc: [EMAIL PROTECTED] Subject: Re: [R] Smooth monotone estimation on R help.search() on my machine turns up only: mono.con(mgcv) Monotonicity constraints for a cubic regression spline. smooth.monotone(fda)Monotone Smoothing of Data pmreg(ftnonpar) Piecewise monotone regression with taut strings backSpline(splines) Monotone Inverse Spline isoreg(stats) Isotonic / Monotone Regression so you should find something of use in packages mgvc, fda, ftnonpar, splines or stats (.loaded by default) Kjetil Halvorsen Assaf P Oron wrote: Hi all, I'm looking for smooth monotone estimation packages, preferably using splines. I downloaded the 'cobs' package and intend to use it, but since it offers only quadratic splines based on L1 minimization, I'd like to compare its performance to that of a more 'mainstream' cubic-spline, L2-norm minimizing spline. Preferably a smoothing spline. Does anyone know of such code existing anywhere? Or another smooth monotone alternative? Thanks in advance, Assaf Oron Statistics Department University of Washington __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] constrOptim and function with additional parameters?
Hi Marlene,
from the on-line help file of `constrOptim' you can see that the ...
argument is used for passing extra arguments to the `optim' function
and not in the function being optimized under constraints.
A simple solution would be to pass the value of the extra argument
directly to the function definition e.g., `fr(x, a=100)'. Otherwise
you could create a new version of `constrOptim' that passes the ...
to the function being optimized and `optim'
I hope this helps.
Best,
Dimitris
Dimitris Rizopoulos
Doctoral Student
Biostatistical Centre
School of Public Health
Catholic University of Leuven
Address: Kapucijnenvoer 35, Leuven, Belgium
Tel: +32/16/396887
Fax: +32/16/337015
Web: http://www.med.kuleuven.ac.be/biostat/
http://www.student.kuleuven.ac.be/~m0390867/dimitris.htm
- Original Message -
From: Marlene Mueller [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Sent: Wednesday, July 14, 2004 2:59 PM
Subject: [R] constrOptim and function with additional parameters?
How can I use a function with some additional input parameters
in constrOptim? For example, something like
fr - function(x,a) { ## Rosenbrock Banana function
x1 - x[1]
x2 - x[2]
a * (x2 - x1 * x1)^2 + (1 - x1)^2
}
where the optimum is to be found w.r.t. x. Calling
optim(c(-1.2,1), fr, NULL, a=100) works as expected, but I fail
to provide the a=100 in the constrained case:
constrOptim(c(-1.2,0.9), fr, NULL, ui=rbind(c(-1,0),c(0,-1)),
ci=c(-1,-1),a=100)
Error in f(theta) : Argument a is missing, with no default
Is this a bug or is there a different solution that I miss here?
TIA, Marlene
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] constrOptim and function with additional parameters?
On Wed, 14 Jul 2004 14:59:01 +0200 (MEST), Marlene Mueller
[EMAIL PROTECTED] wrote :
How can I use a function with some additional input parameters
in constrOptim? For example, something like
fr - function(x,a) { ## Rosenbrock Banana function
x1 - x[1]
x2 - x[2]
a * (x2 - x1 * x1)^2 + (1 - x1)^2
}
where the optimum is to be found w.r.t. x. Calling
optim(c(-1.2,1), fr, NULL, a=100) works as expected, but I fail
to provide the a=100 in the constrained case:
constrOptim(c(-1.2,0.9), fr, NULL, ui=rbind(c(-1,0),c(0,-1)),
ci=c(-1,-1),a=100)
Error in f(theta) : Argument a is missing, with no default
Is this a bug or is there a different solution that I miss here?
I can't spot why your use of constrOptim isn't working, but you should
be able to workaround it by doing something like this:
applyDefaults - function(fn, ...) {
function(x) fn(x, ...)
}
constrOptim(c(-1.2,0.9), applyDefaults(fr, a=100), NULL,
ui=rbind(c(-1,0),c(0,-1)),ci=c(-1,-1))
The applyDefaults function creates a new function which evaluates the
old one with some of the parameters set to fixed values.
Duncan Murdoch
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] constrOptim and function with additional parameters?
Actually, I think this is a bug. Take a look at this part of constrOptim:
constrOptim
function (theta, f, grad, ui, ci, mu = 1e-04, control = list(),
method = if (is.null(grad)) Nelder-Mead else BFGS,
outer.iterations = 10
0,
outer.eps = 1e-05, ...)
{
if (!is.null(control$fnscale) control$fnscale 0)
mu - -mu
[...]
obj - f(theta)
^^^
r - R(theta, theta)
for (i in 1:outer.iterations) {
obj.old - obj
r.old - r
[...]
}
So the object function `f' is called on the starting value `theta' but
the `...' is not passed through.
-roger
Duncan Murdoch wrote:
On Wed, 14 Jul 2004 14:59:01 +0200 (MEST), Marlene Mueller
[EMAIL PROTECTED] wrote :
How can I use a function with some additional input parameters
in constrOptim? For example, something like
fr - function(x,a) { ## Rosenbrock Banana function
x1 - x[1]
x2 - x[2]
a * (x2 - x1 * x1)^2 + (1 - x1)^2
}
where the optimum is to be found w.r.t. x. Calling
optim(c(-1.2,1), fr, NULL, a=100) works as expected, but I fail
to provide the a=100 in the constrained case:
constrOptim(c(-1.2,0.9), fr, NULL, ui=rbind(c(-1,0),c(0,-1)),
ci=c(-1,-1),a=100)
Error in f(theta) : Argument a is missing, with no default
Is this a bug or is there a different solution that I miss here?
I can't spot why your use of constrOptim isn't working, but you should
be able to workaround it by doing something like this:
applyDefaults - function(fn, ...) {
function(x) fn(x, ...)
}
constrOptim(c(-1.2,0.9), applyDefaults(fr, a=100), NULL,
ui=rbind(c(-1,0),c(0,-1)),ci=c(-1,-1))
The applyDefaults function creates a new function which evaluates the
old one with some of the parameters set to fixed values.
Duncan Murdoch
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] (no subject)
Hello, I'm new to R, and I'm having trouble importing a text file (I'm on Windows XP) m - read.table(/Desktop/work/128_L) Error in file(file, r) : unable to open connection In addition: Warning message: cannot open file `/Desktop/work/128_L' do you know why this isn't working? All I have is a bunch of text files, each with a single column of about 30,000 rows. thanks [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Smooth monotone estimation on R
Andy, Kjetil, hi,
I figured out how to run this, after visiting Jim Ramsay's FDA webpage
example on monotone smoothing:
http://ego.psych.mcgill.ca/misc/fda/ex-growth-d1.html
The code there is Matlab, using very similar names to the R names.
First one needs to create a 'blank' B-spline object (using
create.bspline.basis(), then data2fd() with a 'blank' data vector of, say,
zeros).
Then this object is used as the 'Wfdobj' argument in 'smooth.monotone()'.
The smoothing spline version (lambda=some positive constant) seems to work
better than non-penalized splines (lambda=0).
So the command would be
solution-smooth.monotone(x,y,Wfdobj=createdbspline,lambda=0.1)
The fit may be shown using
fitted-solution$beta[1]+solution$beta[2]*eval.monfd(xx,solution$Wfdobj).
eval.monfd() is a utility to perform the integration on the exponent of the
'embedded' B-spline object.
Algebraic details do appear in the 'smooth.monotone()' help, and also on the
webpage.
Ramsay's approach seems to be the one most 'popular' in literature at the
moment, so I'll probably use this package a lot over the next few weeks. If
someone can't access Jim Ramsay or some other 'FDA insider', you can
probably forward them to me regarding smooth.monotone().
Thanks so much again,
Assaf
-Original Message-
From: Liaw, Andy [mailto:[EMAIL PROTECTED]
Sent: Wednesday, July 14, 2004 6:17 AM
To: 'Kjetil Halvorsen'; Eliyahu-Oron
Cc: [EMAIL PROTECTED]
Subject: RE: [R] Smooth monotone estimation on R
Browsing over the FDA book, I can not find any discussion of monotone
smoothing.
Andy
From: Kjetil Halvorsen
Hola!
Experimenting a little, package fda seems very much under
development,
and help pages are definitely not finished.
It would certainly help to read the book functional data
analysis by
the author of fda, which is in our library.
On the other hand package mgcv is more mature, so it would
seem easier
to use that.
It is not totally automatic, but the example
library(mgcv)
example(mono.con)
has everything necessary to get started.
Kjetil Halvorsen
Eliyahu-Oron wrote:
Kjetil and Andy,
Thanks for your helpful answers! The first two (mgcv and
fda) seem to be in
the direction I'm looking for. I downloaded them both.
I'm running into a lot of implementation difficulties,
though. I wonder if
there's anyone who tried to do a monotone spline using
either the 'mgcv' or
the 'fda' packages, whom I could ask directly?
Thanks again, Assaf
-Original Message-
From: Kjetil Halvorsen [mailto:[EMAIL PROTECTED]
Sent: Monday, July 12, 2004 4:32 PM
To: Assaf P Oron
Cc: [EMAIL PROTECTED]
Subject: Re: [R] Smooth monotone estimation on R
help.search()
on my machine turns up only:
mono.con(mgcv) Monotonicity constraints for a cubic
regression spline.
smooth.monotone(fda)Monotone Smoothing of Data
pmreg(ftnonpar) Piecewise monotone regression with taut
strings
backSpline(splines) Monotone Inverse Spline
isoreg(stats) Isotonic / Monotone Regression
so you should find something of use in packages mgvc, fda,
ftnonpar, splines
or stats (.loaded by default)
Kjetil Halvorsen
Assaf P Oron wrote:
Hi all,
I'm looking for smooth monotone estimation packages,
preferably using
splines.
I downloaded the 'cobs' package and intend to use it, but
since it offers
only quadratic splines based on L1 minimization, I'd like to
compare its
performance to that of a more 'mainstream' cubic-spline,
L2-norm minimizing
spline. Preferably a smoothing spline.
Does anyone know of such code existing anywhere? Or another
smooth monotone
alternative?
Thanks in advance,
Assaf Oron
Statistics Department
University of Washington
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
--
Notice: This e-mail message, together with any attachments,...{{dropped}}
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] (no subject)
Hi, I but you left out the extension that Windows is hiding from you. Vincent On Wednesday 14 July 2004 10:02, Herman, David (NIH/NIMH) wrote: Hello, I'm new to R, and I'm having trouble importing a text file (I'm on Windows XP) m - read.table(/Desktop/work/128_L) Error in file(file, r) : unable to open connection In addition: Warning message: cannot open file `/Desktop/work/128_L' do you know why this isn't working? All I have is a bunch of text files, each with a single column of about 30,000 rows. thanks [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Vincent Goulet, Professeur agrégé École d'actuariat Université Laval, Québec [EMAIL PROTECTED] http://vgoulet.act.ulaval.ca __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] (no subject)
Hello! It just is not able to find the file as the error message says. You can check if you are specified the right directory path using dir. ?dir dir(/Desktop/work/) I gues you are not. If you are specifying the full path, precede it with the drive name c:/ Eryk *** REPLY SEPARATOR *** On 7/14/2004 at 10:02 AM Herman, David (NIH/NIMH) wrote: Hello, I'm new to R, and I'm having trouble importing a text file (I'm on Windows XP) m - read.table(/Desktop/work/128_L) Error in file(file, r) : unable to open connection In addition: Warning message: cannot open file `/Desktop/work/128_L' do you know why this isn't working? All I have is a bunch of text files, each with a single column of about 30,000 rows. thanks [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Dipl. bio-chem. Eryk Witold Wolski@MPI-Moleculare Genetic Ihnestrasse 63-73 14195 Berlin 'v' tel: 0049-30-83875219 / \ mail: [EMAIL PROTECTED]---W-Whttp://www.molgen.mpg.de/~wolski __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] ROracle - fetch gives empty dataframe
This is what I would try next if I were in your situation; I don't know it will help. Try con - dbConnect(ora, rado/[EMAIL PROTECTED]) instead of con - dbConnect(ora, rado/only2admin The default dbname is Sys.getenv(ORACLE_SID), have you checked that? For myself, when I make connections to Oracle 9i using ROracle, I use the form dbConnect,ora, user='oracleuid',dbname='adbname',password='whatever') -Don At 10:38 AM +0200 7/14/04, Rado Bonk wrote: Dear R-users, I was able to make ROracle package to connect to the DB (Oracle91, 64bit, on Solaris). But after executing siple SQL query, fetch commaned gives me an empty dataframe. ### RORACLE INSTALATION PROCEDURE ### R CMD INSTALL --configure-args='--enable-extralibs' --enable-oracle32=no ~/tmp/ROracle_0.5-5.tar.gz #since we have 64bit Oracle9i instalation PROBLEM: after executing simple statement, from within R, fetch function gives me the empty dataframe: ### CONNECTION, and SQL QUERY EXECUTION ora - dbDriver(Oracle) con - dbConnect(ora, rado/only2admin dbListTables(con) character(0) rs - dbSendQuery(con, desc * from si_r where id=498) d - fetch(rs, n= -1) dim(d) [1] 0 3 str(d) data.frame': 0 obs. of 3 variables: $ ID : int $ DAY: chr $ R : num seems like connections has been established dbGetInfo(ora) $drvName [1] Oracle (ProC/C++) $connectionIds $connectionIds[[1]] OraConnection:(14939,0) $fetch_default_rec [1] 500 $managerId OraDriver:(14939) $length [1] 10 $num_con [1] 1 $counter [1] 1 $clientVersion [1] 0.5-4 Maybe it is something stupid (and I need just a hint), or may be it is something major. Thanks in advance, Rado Bonk -- Dr. Radoslav Bonk European Commission - DG Joint Research Centre (JRC) Institute for Environment and Sustainability (IES) LM Unit - Natural Hazards Weather Driven Natural Hazards Action Via E. Fermi, TP 261, 21020 Ispra (Va), Italy Tel.: 0039-0332-786013 Fax: 0039-0332-786653 Webpage: http://natural-hazards.jrc.it/floods/ __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- -- Don MacQueen Environmental Protection Department Lawrence Livermore National Laboratory Livermore, CA, USA __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] (no subject)
On Wed, 14 Jul 2004 10:11:30 -0400, Vincent Goulet [EMAIL PROTECTED] wrote : Hi, I but you left out the extension that Windows is hiding from you. Yes indeed! Why Windows hides extensions has never made any sense to me. To turn off this seriously broken behaviour, open a folder, then in the Tools|Folder Options dialog, find Hide extensions for known file types and make sure it is not checked. In Win XP it's under the View tab, around 10 lines down; I think it's in other places in other versions. Duncan Murdoch Vincent On Wednesday 14 July 2004 10:02, Herman, David (NIH/NIMH) wrote: Hello, I'm new to R, and I'm having trouble importing a text file (I'm on Windows XP) m - read.table(/Desktop/work/128_L) Error in file(file, r) : unable to open connection In addition: Warning message: cannot open file `/Desktop/work/128_L' do you know why this isn't working? All I have is a bunch of text files, each with a single column of about 30,000 rows. thanks [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Running the optimization on the subset of parameters
Dear all, I'd like to find a minimum of (-loglik) function which is a function of k parameters. I'd like to run the minimization algorithm for the different subsets of the parameters and assign the fixed values to the complementary subset. How should I define my (-loglik) function such that it can be passed to the optim or other optimization function? Much thanks for any suggestions. Vicky Landsman. [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] table lookup n R
There is also the match() function, and the %in% operator, either of
which might do the job, depending on your exact details. For example,
(1:26)[letters %in% c('x','t','j')]
-Don
At 2:34 PM +0200 7/13/04, Anne wrote:
Hello R helpers!
I looked but did not find a table-lookup R-utility. I could use a
loop to do the job (old FORTRAN/C habits die hard) but if I have a
big table in which I have to search for the values corresponding to
a vector, I end up logically with a double loop.
Is there already such a utility? Otherwise, is there a way without loops?
Thanks as always
Anne
Anne Piotet
Tel: +41 79 359 83 32 (mobile)
Email: [EMAIL PROTECTED]
---
M-TD Modelling and Technology Development
PSE-C
CH-1015 Lausanne
Switzerland
Tel: +41 21 693 83 98
Fax: +41 21 646 41 33
--
[[alternative HTML version deleted]]
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
--
--
Don MacQueen
Environmental Protection Department
Lawrence Livermore National Laboratory
Livermore, CA, USA
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] constrOptim and function with additional parameters?
This is why the `...' argument is passed to `optim' and not to the
function being optimized, as it also referred in the help file of
`constrOptim'
constrOptim
function (theta, f, grad, ui, ci, mu = 1e-04, control = list(),
method = if (is.null(grad)) Nelder-Mead else BFGS,
outer.iterations = 100,
outer.eps = 1e-05, ...)
...
a - optim(theta.old, fun, gradient, control = control,
method = method, ...)
...
}
Best,
Dimitris
Dimitris Rizopoulos
Doctoral Student
Biostatistical Centre
School of Public Health
Catholic University of Leuven
Address: Kapucijnenvoer 35, Leuven, Belgium
Tel: +32/16/396887
Fax: +32/16/337015
Web: http://www.med.kuleuven.ac.be/biostat/
http://www.student.kuleuven.ac.be/~m0390867/dimitris.htm
- Original Message -
From: Roger D. Peng [EMAIL PROTECTED]
To: Duncan Murdoch [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]; Marlene Mueller
[EMAIL PROTECTED]
Sent: Wednesday, July 14, 2004 4:01 PM
Subject: Re: [R] constrOptim and function with additional parameters?
Actually, I think this is a bug. Take a look at this part of
constrOptim:
constrOptim
function (theta, f, grad, ui, ci, mu = 1e-04, control = list(),
method = if (is.null(grad)) Nelder-Mead else BFGS,
outer.iterations = 10
0,
outer.eps = 1e-05, ...)
{
if (!is.null(control$fnscale) control$fnscale 0)
mu - -mu
[...]
obj - f(theta)
^^^
r - R(theta, theta)
for (i in 1:outer.iterations) {
obj.old - obj
r.old - r
[...]
}
So the object function `f' is called on the starting value `theta'
but
the `...' is not passed through.
-roger
Duncan Murdoch wrote:
On Wed, 14 Jul 2004 14:59:01 +0200 (MEST), Marlene Mueller
[EMAIL PROTECTED] wrote :
How can I use a function with some additional input parameters
in constrOptim? For example, something like
fr - function(x,a) { ## Rosenbrock Banana function
x1 - x[1]
x2 - x[2]
a * (x2 - x1 * x1)^2 + (1 - x1)^2
}
where the optimum is to be found w.r.t. x. Calling
optim(c(-1.2,1), fr, NULL, a=100) works as expected, but I fail
to provide the a=100 in the constrained case:
constrOptim(c(-1.2,0.9), fr, NULL, ui=rbind(c(-1,0),c(0,-1)),
ci=c(-1,-1),a=100)
Error in f(theta) : Argument a is missing, with no default
Is this a bug or is there a different solution that I miss here?
I can't spot why your use of constrOptim isn't working, but you
should
be able to workaround it by doing something like this:
applyDefaults - function(fn, ...) {
function(x) fn(x, ...)
}
constrOptim(c(-1.2,0.9), applyDefaults(fr, a=100), NULL,
ui=rbind(c(-1,0),c(0,-1)),ci=c(-1,-1))
The applyDefaults function creates a new function which evaluates
the
old one with some of the parameters set to fixed values.
Duncan Murdoch
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] notes on specifying paths and file extensions in Windows; was: (no subject)
Duncan Murdoch wrote: On Wed, 14 Jul 2004 10:11:30 -0400, Vincent Goulet [EMAIL PROTECTED] wrote : Hi, I but you left out the extension that Windows is hiding from you. Yes indeed! Why Windows hides extensions has never made any sense to me. To turn off this seriously broken behaviour, open a folder, then in the Tools|Folder Options dialog, find Hide extensions for known file types and make sure it is not checked. In Win XP it's under the View tab, around 10 lines down; I think it's in other places in other versions. In particular, the path for sure does not realy start with /Desktop... You might want to try dir(file.path(Sys.getenv(USERPROFILE), Desktop/work)) and correct the entry to read.table(file.path(Sys.getenv(USERPROFILE), Desktop/work/128_L.YourExtension)) Uwe Ligges BTW: I wonder why nobody uses a sensible subject in this thread Duncan Murdoch Vincent On Wednesday 14 July 2004 10:02, Herman, David (NIH/NIMH) wrote: Hello, I'm new to R, and I'm having trouble importing a text file (I'm on Windows XP) m - read.table(/Desktop/work/128_L) Error in file(file, r) : unable to open connection In addition: Warning message: cannot open file `/Desktop/work/128_L' do you know why this isn't working? All I have is a bunch of text files, each with a single column of about 30,000 rows. thanks [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] ROracle - fetch gives empty dataframe
Michael, Yes you are right, I forgot to specify the dbname when posting in R-help list. But not in my R-code, since it is specified in ORACLE_SID variable on Linux. Using the proper syntax I still get an empty data frame: library(ROracle) drv - dbDriver(Oracle) con - dbConnect(drv, rado/[EMAIL PROTECTED]) dbListTables(con) character(0) q.sql - c(select * from si_r where id=498) rs - dbSendQuery(con, q.sql) data - fetch(rs, n= -1) dim(data) [1] 0 3 dbGetStatement(rs) [1] select * from si_r where id=498 dbGetInfo(con) $dbname [1] dea $user [1] rado $passwd [1] mypassword $conType [1] NA $serverVersion [1] NA $protocolVersion [1] NA $threadId [1] -1 $resultSetIds $resultSetIds[[1]] OraResult:(15551,0,2) data [1] ID DAY R 0 rows (or 0-length row.names) ### CONNECTION, and SQL QUERY EXECUTION ora - dbDriver(Oracle) con - dbConnect(ora, rado/only2admin Which database??? -- Radoslav Bonk European Commission - DG Joint Research Centre (JRC) Institute for Environment and Sustainability (IES) LM Unit - Natural Hazards Weather Driven Natural Hazards Action Via E. Fermi, TP 261, 21020 Ispra (Va), Italy Tel.: 0039-0332-786013 Fax: 0039-0332-786653 Webpage: http://natural-hazards.jrc.it/floods/ __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] duplicate row importing issue
Hello, I'm simply trying to import a .txt file that has a column of about 30,000 pts. I found the file, but I'm getting an error: m - read.table(choose.files()) Error in row.names-.data.frame(`*tmp*`, value = row.names) : duplicate row.names are not allowed Any help with getting around this? I really appreciate all the help. Thanks dave [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Tcl/Tk and R
Hi!! I'm using the R package tcltk. See the stretch of code below: tbn1 = tclvalue(tkadd(tn,label=Modify Data)) tkpack(tbw1 - .Tk.newwin(tbn1)) tkpack(fr1 - tkframe(tbw1)) tkpack(lb1- tkwidget(fr1,iwidgets::labeledframe)) When I try to execute this I have this error at the last line: Error in structure(.External(dotTclObjv, objv, PACKAGE = tcltk), class = tclObj) : [tcl] can't find usual code for tag Frame. Please help me!! Thank's Talita Perciano Costa Leite Graduanda em Ciência da Computação Universidade Federal de Alagoas - UFAL Departamento de Tecnologia da Informação - TCI Construção de Conhecimento por Agrupamento de Dados - CoCADa __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] duplicate row importing issue
Try read.table(choose.files(), row.names=NULL). BTW, I think you might be using an older R version because in R-1.9.1, the value for row.names is missing by default in read.table(). args(read.table) function (file, header = FALSE, sep = , quote = \', dec = ., row.names, col.names, as.is = FALSE, na.strings = NA, colClasses = NA, nrows = -1, skip = 0, check.names = TRUE, fill = !blank.lines.skip, strip.white = FALSE, blank.lines.skip = TRUE, comment.char = #) Try read.delim() or read.csv() if the data is tab-delimited or comma separated. Also check to see if you have a header, in which case you would set the option header=TRUE. For more info, help(read.table). Do dim(m), head(m) to see if the data has been read correctly. On Wed, 2004-07-14 at 16:12, Herman, David (NIH/NIMH) wrote: Hello, I'm simply trying to import a .txt file that has a column of about 30,000 pts. I found the file, but I'm getting an error: m - read.table(choose.files()) Error in row.names-.data.frame(`*tmp*`, value = row.names) : duplicate row.names are not allowed Any help with getting around this? I really appreciate all the help. Thanks dave [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] duplicate row importing issue
Dear David, If there is one fewer variable name in the first row of the data file than fields in the remaining rows, then read.table() will treat the first entry in each row as the row name. Simply add a new first variable (such as name) to the first row, and specify header=TRUE in the call to read.table(). (Of course, this assumes that I've correctly guessed the source of the problem.) I hope that this helps, John -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Herman, David (NIH/NIMH) Sent: Wednesday, July 14, 2004 10:13 AM To: '[EMAIL PROTECTED]' Subject: [R] duplicate row importing issue Hello, I'm simply trying to import a .txt file that has a column of about 30,000 pts. I found the file, but I'm getting an error: m - read.table(choose.files()) Error in row.names-.data.frame(`*tmp*`, value = row.names) : duplicate row.names are not allowed Any help with getting around this? I really appreciate all the help. Thanks dave __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] SWIG for R
On Wed, 14 Jul 2004 08:01:28 -0700, Duncan Temple Lang [EMAIL PROTECTED] wrote : SWIG is an extensible system and so people other than the SWIG developers can indeed provide facilities for supporting R. I am surprised nobody has done it yet and remember asking you whether you had considered using SWIG for your OpenGL about 3 years ago. Sorry, I forgot about that. I don't think SWIG would be much help with OpenGL, because there the difficulty is in translating the R ideas of what data is like and what people want to do with it into corresponding concepts in OpenGL, the actual calls to the OpenGL API are a pretty easy part of the whole exercise. In any case, they probably have to be written in compiled code for performance reasons, since you make so many OpenGL calls for every frame being drawn. But I'm sure there are other libraries where this isn't true, and SWIG would be useful for them. Maybe Hisaji has one, and would want to tackle the R support. Duncan Murdoch __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Permutations
Dear R users,
First of all, thanks to Rolf, Brad, Robin, Erich, Fernando and Adaikalavan
for your time and suggestions.
Ive been testing some algorithms (sorry for the delay, Im very slow, and
Im a completely beginner in Rs world).
First, the Robin algorithm.
I think that there is a problem because Ive done 200 permutations and
Ive found that these permutations are the same:
52 and 91, 99 and 110, 121 and 122, 51 and 141, 130 and 134.
Thanks again,
Jordi Altirriba
Hospital Clinic Barcelona - Spain
x - c(1,2,3,4,5,6,7,8,9,10,11,12)
dim(x) - c(3,4) a-matrix(1,200,12)
for (i in 1:200)
+ {
+ jj - t(apply(x,1,sample))
+ a[i,]-as.vector(jj)
+ }
a
[,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10] [,11] [,12]
[1,]7231 11648910 512
[2,]1297 11648 1210 5 3
[3,]7291 11345610 812
[4,] 108645 12729 111 3
[5,] 102 121 119786 4 5 3
[6,]7861594 11 1210 2 3
[7,]15 127264891011 3
[8,]159 1086423 71112
[9,]1 11672 1245910 8 3
[10,]45 12 10 119186 7 2 3
[11,]1 11975648 1210 2 3
[12,]18342 12 1059 711 6
[13,]1237 116 105 12 4 8 9
[14,]483 105 12729 111 6
[15,] 10234867 119 1 512
[16,]489 102 12756 111 3
[17,]126 105378 12 411 9
[18,] 10294 11 12156 7 8 3
[19,]4867 11 1212910 5 3
[20,]18 12723 10 116 4 5 9
[21,] 102 121597 116 4 8 3
[22,]4 11 12123 1086 7 5 9
[23,]1 113726 1059 4 812
[24,]729 105 121 113 4 8 6
[25,]789126453101112
[26,]45 12 10237 116 1 8 9
[27,]4591 11378 1210 2 6
[28,]1564 11378910 212
[29,]4561 119 102 12 7 8 3
[30,]4 11378 12 1056 1 2 9
[31,] 10231 116789 4 512
[32,] 10237891 116 4 512
[33,]7 11618945 1210 2 3
[34,]75 121864 11310 2 9
[35,]123486759101112
[36,]7831 119 102 12 4 5 6
[37,] 10261 11 12753 4 8 9
[38,]1594 11 1278310 2 6
[39,]12 12759 1083 411 6
[40,]183 102 127 116 4 5 9
[41,]129483 10 11 12 7 5 6
[42,]456129 1083 71112
[43,]1267 11 12 1059 4 8 3
[44,]129 10 11 12486 7 5 3
[45,] 10597 116423 1 812
[46,]1234 11675910 812
[47,]426183 105 12 711 9
[48,]48972315 121011 6
[49,] 108 121294 113 7 5 6
[50,] 108612375 12 411 9
[51,]72 12 10 116483 1 5 9
[52,]45612 12 10 119 7 8 3
[53,]123756489101112
[54,] 10537 119186 4 212
[55,]7 11 12423 1086 1 5 9
[56,]1594 11 12 1083 7 2 6
[57,]4597 113 1026 1 812
[58,] 10 11345618 12 7 2 9
[59,]489 1056723 11112
[60,]42 12186 1059 711 3
[61,]4867 119 105 12 1 2 3
[62,]783 10561 11 12
Re: [R] SWIG for R
I like SWIG (well, it facilitates wrappers for Python really well),
but it's not perfect.
I've not delved into the difficulties in any detail yet, but we've had
problems with mixing SWIG and Boost.python for connections with Python
code -- and the Boost.python mechanism feels like DTL's RSyour
language here code, where as SWIG doesn't quite have the same flavor.
(sorry for failing on the details, but I'm doing something else
today).
Duncan Murdoch [EMAIL PROTECTED] writes:
On Wed, 14 Jul 2004 08:01:28 -0700, Duncan Temple Lang
[EMAIL PROTECTED] wrote :
SWIG is an extensible system and so people
other than the SWIG developers can indeed
provide facilities for supporting R.
I am surprised nobody has done it yet
and remember asking you whether you had considered
using SWIG for your OpenGL about 3 years ago.
Sorry, I forgot about that.
I don't think SWIG would be much help with OpenGL, because there the
difficulty is in translating the R ideas of what data is like and what
people want to do with it into corresponding concepts in OpenGL, the
actual calls to the OpenGL API are a pretty easy part of the whole
exercise. In any case, they probably have to be written in compiled
code for performance reasons, since you make so many OpenGL calls
for every frame being drawn.
But I'm sure there are other libraries where this isn't true, and SWIG
would be useful for them. Maybe Hisaji has one, and would want to
tackle the R support.
Duncan Murdoch
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
--
Anthony Rossini Research Associate Professor
[EMAIL PROTECTED]http://www.analytics.washington.edu/
Biomedical and Health Informatics University of Washington
Biostatistics, SCHARP/HVTN Fred Hutchinson Cancer Research Center
UW (Tu/Th/F): 206-616-7630 FAX=206-543-3461 | Voicemail is unreliable
FHCRC (M/W): 206-667-7025 FAX=206-667-4812 | use Email
CONFIDENTIALITY NOTICE: This e-mail message and any attachme...{{dropped}}
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Ord-Getis O statistics
Hi list, I am wondering if anybody knows if the Ord-Getis O statistics of local spatial autocorrelation in the presence of the global spatial association is implemented in any of the R packages - and of course in which package ;-)). I am not interested in Getis-Ord G statistics, for now. Thank you in advance, Monica Monica Palaseanu-Lovejoy University of Manchester School of Geography Mansfield Cooper Bld. 3.21 Oxford Road Manchester M13 9PL England, UK Tel: +44 (0) 275 8689 Email: [EMAIL PROTECTED] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Permutations
As Erich points out, there is some question as to what the original problem really is but lets assume its as Erich describes. Then, to get a random ordered permutation we just get a random permutation of 12 elements and sort the intra-block elements like this: c(apply(matrix(sample(12,12),3),2,sort)) Note that there are 12! = 479,001,600 permutations of 12 elements and 3!^4 = 1,296 of those permutations correspond to each ordered permutation so there are 479,001,600 / 1,296 = 369,600 ordered permutations in all. Erich Neuwirth erich.neuwirth at univie.ac.at writes: : : Perhaps what you want might better be described as : ordered partitions? : : Is what you want the following: : : We study sequences of length 12 and divide them in : 4 segments : position 1 2 3, position 4 5 6, : position 7 8 9, position 10 11 12, : : Find all permutation sequences of the numbers 1 to 12 : with the property that all segment sequences : are monotonically increasing. : : I think that produces what you need. : Since the segments are ordered, you avoid intra-block permutations. : : If that is what you want, writing a recursive function should not be : too hard. : : Jordi Altirriba Gutirrez wrote: : : Dear R users, : Im a beginner user of R and Ive a problem with permutations that I : dont know how to solve. Ive 12 elements in blocks of 3 elements and I : want only to make permutations inter-blocks (no intra-blocks) (sorry if : the terminology is not accurate), something similar to: : : 1 2 3 | 4 5 6 | 7 8 9 | 10 11 12 --1st permutation : : 1 3 2 | 4 5 6 | 7 8 9 | 10 11 12 NO :- - : 3 2 1 | 4 5 6 | 7 8 9 | 10 11 12 NO : - - - : 1 2 4 | 3 5 6 | 7 8 9 | 10 11 12 YES-2nd permutation : -- : 4 5 6 | 1 2 3 | 7 8 9 | 10 11 12 YES-3rd permutation : - - - - - - : 4 5 6 | 2 1 3 | 7 8 9 | 10 11 12 NO :- - : : : Thanks for your time, : : Jordi Altirriba : Ph D student : : Hospital Clinic Barcelona - Spain : : : MSN Motor. http://motor.msn.es/researchcentre/ : : __ : R-help at stat.math.ethz.ch mailing list : https://www.stat.math.ethz.ch/mailman/listinfo/r-help : PLEASE do read the posting guide! : http://www.R-project.org/posting-guide.html : : __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] SWIG for R
[EMAIL PROTECTED] (A.J. Rossini) writes:
I like SWIG (well, it facilitates wrappers for Python really well),
but it's not perfect.
I've not delved into the difficulties in any detail yet, but we've had
problems with mixing SWIG and Boost.python for connections with Python
code -- and the Boost.python mechanism feels like DTL's RSyour
language here code, where as SWIG doesn't quite have the same flavor.
(sorry for failing on the details, but I'm doing something else
today).
Argh. To be clear, this is in the context of C++ systems. Which is
relevant but completely different.
--
Anthony Rossini Research Associate Professor
[EMAIL PROTECTED]http://www.analytics.washington.edu/
Biomedical and Health Informatics University of Washington
Biostatistics, SCHARP/HVTN Fred Hutchinson Cancer Research Center
UW (Tu/Th/F): 206-616-7630 FAX=206-543-3461 | Voicemail is unreliable
FHCRC (M/W): 206-667-7025 FAX=206-667-4812 | use Email
CONFIDENTIALITY NOTICE: This e-mail message and any attachme...{{dropped}}
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Is there a statistics that can summarize the correlation for more than two random variables?
Thank you for your reminding. Could you tell me the addresses of STAT-L and ALLSTAT lists? By the way, I found Cronbach's alpha suggested by Prof. Baron might be the one I am looking for though it's not perfect. Frank -Original Message- From: Peter Flom [mailto:[EMAIL PROTECTED] Sent: Tuesday, July 13, 2004 16:06 To: [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: Re: [R] Is there a statistics that can summarize the correlation formore than two random variables? This seems more like a STATS question than an R question - asking on a list like STAT-L or ALLSTAT may result in more replies Nevertheless, it seems to me that you need to describe (and maybe decide) what you mean by 'summarize' the correlations. Certainly the mean DOES summarize them, but is it the summary you want? Maybe, maybe not. Perhaps the median? Or a trimmed mean? Do you want to take the absolute values of the correlations, or not? HTH Peter L. Flom, PhD Assistant Director, Statistics and Data Analysis Core Center for Drug Use and HIV Research National Development and Research Institutes 71 W. 23rd St www.peterflom.com New York, NY 10010 (212) 845-4485 (voice) (917) 438-0894 (fax) F Duan [EMAIL PROTECTED] 07/13/04 2:34 PM Hi, R people, I wonder if there is a statistics than can measure the correlation for more than two random variables, instead of computing the correlation coefficient matrix. If so, what R package should I use? Right now I can only think of the mean of all pair-wise correlation coefficients, e.g., (corr(x,y) + corr(x,z) + corr(y,z)) / 3 for three random variables (x, y, z). Thanks a lot, Frank [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] constrOptim and function with additional parameters?
OK, I got the point why the help refers to the arguments
of 'optim' only. (Of course, I have read this but did'nt
believe it ... ;-))
Thanks to all of you for the clarification and special thanks
to Duncan Murdoch for the workaround, which is very close to
what I want. I am doing a lot of constrained regression in the
moment, so including additional parameters (=the data) is an
issue for me.
Best regards, Marlene
This is why the `...' argument is passed to `optim' and not to the
function being optimized, as it also referred in the help file of
`constrOptim'
constrOptim
function (theta, f, grad, ui, ci, mu = 1e-04, control = list(),
method = if (is.null(grad)) Nelder-Mead else BFGS,
outer.iterations = 100,
outer.eps = 1e-05, ...)
...
a - optim(theta.old, fun, gradient, control = control,
method = method, ...)
...
}
Best,
Dimitris
Dimitris Rizopoulos
Doctoral Student
Biostatistical Centre
School of Public Health
Catholic University of Leuven
Address: Kapucijnenvoer 35, Leuven, Belgium
Tel: +32/16/396887
Fax: +32/16/337015
Web: http://www.med.kuleuven.ac.be/biostat/
http://www.student.kuleuven.ac.be/~m0390867/dimitris.htm
- Original Message -
From: Roger D. Peng [EMAIL PROTECTED]
To: Duncan Murdoch [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]; Marlene Mueller
[EMAIL PROTECTED]
Sent: Wednesday, July 14, 2004 4:01 PM
Subject: Re: [R] constrOptim and function with additional parameters?
Actually, I think this is a bug. Take a look at this part of
constrOptim:
constrOptim
function (theta, f, grad, ui, ci, mu = 1e-04, control = list(),
method = if (is.null(grad)) Nelder-Mead else BFGS,
outer.iterations = 10
0,
outer.eps = 1e-05, ...)
{
if (!is.null(control$fnscale) control$fnscale 0)
mu - -mu
[...]
obj - f(theta)
^^^
r - R(theta, theta)
for (i in 1:outer.iterations) {
obj.old - obj
r.old - r
[...]
}
So the object function `f' is called on the starting value `theta'
but
the `...' is not passed through.
-roger
Duncan Murdoch wrote:
On Wed, 14 Jul 2004 14:59:01 +0200 (MEST), Marlene Mueller
[EMAIL PROTECTED] wrote :
How can I use a function with some additional input parameters
in constrOptim? For example, something like
fr - function(x,a) { ## Rosenbrock Banana function
x1 - x[1]
x2 - x[2]
a * (x2 - x1 * x1)^2 + (1 - x1)^2
}
where the optimum is to be found w.r.t. x. Calling
optim(c(-1.2,1), fr, NULL, a=100) works as expected, but I fail
to provide the a=100 in the constrained case:
constrOptim(c(-1.2,0.9), fr, NULL, ui=rbind(c(-1,0),c(0,-1)),
ci=c(-1,-1),a=100)
Error in f(theta) : Argument a is missing, with no default
Is this a bug or is there a different solution that I miss here?
I can't spot why your use of constrOptim isn't working, but you
should
be able to workaround it by doing something like this:
applyDefaults - function(fn, ...) {
function(x) fn(x, ...)
}
constrOptim(c(-1.2,0.9), applyDefaults(fr, a=100), NULL,
ui=rbind(c(-1,0),c(0,-1)),ci=c(-1,-1))
The applyDefaults function creates a new function which evaluates
the
old one with some of the parameters set to fixed values.
Duncan Murdoch
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] duplicate row importing issue
How about you delete the first row (column names) and then use header=FALSE in read.table()? Frank -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Herman, David (NIH/NIMH) Sent: Wednesday, July 14, 2004 11:13 To: '[EMAIL PROTECTED]' Subject: [R] duplicate row importing issue Hello, I'm simply trying to import a .txt file that has a column of about 30,000 pts. I found the file, but I'm getting an error: m - read.table(choose.files()) Error in row.names-.data.frame(`*tmp*`, value = row.names) : duplicate row.names are not allowed Any help with getting around this? I really appreciate all the help. Thanks dave [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] reading text file in Windows (was no subject)
John Fox jfox at mcmaster.ca writes: You got several useful suggestions for what you may have done wrong. I often find that it's easier to use read.table(file.choose()) and to navigate to the file in the resulting dialog than to type the path to the file. Related to this, you could issue the command exactly as shown without any arguments at all: file.choose() from the R console, navigate to the correct file and it will return the correct text representation of the filename to use in your read.table. For example, below I navigated to the AUTHORS file in the rw1091 R distribution: R file.choose() [1] C:\\Program Files\\R\\rw1091\\AUTHORS This should help you decipher whether you got the path wrong, the filename wrong, etc. __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Permutations
I think the issue here is in the two keywords - permutations or sample.
AFAIK, permutations should return all admissible (by some rule)
combinations. If this is a large number, as some have pointed out, then
one essentially takes a _sample_ of all admissible combinations. Since
you earlier mentioned that you only want 5-10 outputs, perhaps the
correct term is sampling with restrictions.
There main problem with Robin's method in that all elements within a row
are mutually exclusive to the other. e.g. only one of either 1, 4, 7, 10
can appear in block 1. Furthermore they can only appear in the first
slot of the first block (so no intra-block randomness). This limits the
number of possible outputs.
Can you clearly define the rules (with examples) for an admissible
combination ? They seem to have a different meaning every time I read
the mail. Maybe I am just confused.
On Wed, 2004-07-14 at 17:16, Jordi Altirriba Gutirrez wrote:
Dear R users,
First of all, thanks to Rolf, Brad, Robin, Erich, Fernando and Adaikalavan
for your time and suggestions.
Ive been testing some algorithms (sorry for the delay, Im very slow, and
Im a completely beginner in Rs world).
First, the Robin algorithm.
I think that there is a problem because Ive done 200 permutations and
Ive found that these permutations are the same:
52 and 91, 99 and 110, 121 and 122, 51 and 141, 130 and 134.
Thanks again,
Jordi Altirriba
Hospital Clinic Barcelona - Spain
x - c(1,2,3,4,5,6,7,8,9,10,11,12)
dim(x) - c(3,4) a-matrix(1,200,12)
for (i in 1:200)
+ {
+ jj - t(apply(x,1,sample))
+ a[i,]-as.vector(jj)
+ }
a
[,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10] [,11] [,12]
[1,]7231 11648910 512
[2,]1297 11648 1210 5 3
[3,]7291 11345610 812
[4,] 108645 12729 111 3
[5,] 102 121 119786 4 5 3
[6,]7861594 11 1210 2 3
[7,]15 127264891011 3
[8,]159 1086423 71112
[9,]1 11672 1245910 8 3
[10,]45 12 10 119186 7 2 3
[11,]1 11975648 1210 2 3
[12,]18342 12 1059 711 6
[13,]1237 116 105 12 4 8 9
[14,]483 105 12729 111 6
[15,] 10234867 119 1 512
[16,]489 102 12756 111 3
[17,]126 105378 12 411 9
[18,] 10294 11 12156 7 8 3
[19,]4867 11 1212910 5 3
[20,]18 12723 10 116 4 5 9
[21,] 102 121597 116 4 8 3
[22,]4 11 12123 1086 7 5 9
[23,]1 113726 1059 4 812
[24,]729 105 121 113 4 8 6
[25,]789126453101112
[26,]45 12 10237 116 1 8 9
[27,]4591 11378 1210 2 6
[28,]1564 11378910 212
[29,]4561 119 102 12 7 8 3
[30,]4 11378 12 1056 1 2 9
[31,] 10231 116789 4 512
[32,] 10237891 116 4 512
[33,]7 11618945 1210 2 3
[34,]75 121864 11310 2 9
[35,]123486759101112
[36,]7831 119 102 12 4 5 6
[37,] 10261 11 12753 4 8 9
[38,]1594 11 1278310 2 6
[39,]12 12759 1083 411 6
[40,]183 102 127 116 4 5 9
[41,]129483 10 11 12 7 5 6
[42,]456129 1083 71112
[43,]1267 11 12 1059 4 8 3
[44,]129 10 11 12486 7 5 3
[45,] 10597 116423 1 812
[46,]1234 11675910 812
[47,]426183 105 12 711
RE: [R] Is there a statistics that can summarize the correlation for more than two random variables?
http://www.jiscmail.ac.uk/lists/ALLSTAT.html On Wed, 2004-07-14 at 17:47, F Duan wrote: Thank you for your reminding. Could you tell me the addresses of STAT-L and ALLSTAT lists? By the way, I found Cronbach's alpha suggested by Prof. Baron might be the one I am looking for though it's not perfect. Frank -Original Message- From: Peter Flom [mailto:[EMAIL PROTECTED] Sent: Tuesday, July 13, 2004 16:06 To: [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: Re: [R] Is there a statistics that can summarize the correlation formore than two random variables? This seems more like a STATS question than an R question - asking on a list like STAT-L or ALLSTAT may result in more replies Nevertheless, it seems to me that you need to describe (and maybe decide) what you mean by 'summarize' the correlations. Certainly the mean DOES summarize them, but is it the summary you want? Maybe, maybe not. Perhaps the median? Or a trimmed mean? Do you want to take the absolute values of the correlations, or not? HTH Peter L. Flom, PhD Assistant Director, Statistics and Data Analysis Core Center for Drug Use and HIV Research National Development and Research Institutes 71 W. 23rd St www.peterflom.com New York, NY 10010 (212) 845-4485 (voice) (917) 438-0894 (fax) F Duan [EMAIL PROTECTED] 07/13/04 2:34 PM Hi, R people, I wonder if there is a statistics than can measure the correlation for more than two random variables, instead of computing the correlation coefficient matrix. If so, what R package should I use? Right now I can only think of the mean of all pair-wise correlation coefficients, e.g., (corr(x,y) + corr(x,z) + corr(y,z)) / 3 for three random variables (x, y, z). Thanks a lot, Frank [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Convex smoothing via 'Iterative Convex Minorant' ?
I've been asked, and interested myself:
Has anybody implemented the above in R or another S language dialect?
We are talking about the algorithms / methodology
by Wellner, Groeneboom and Jongbloed, e.g., from the following article
@Article{Jongbloed:1998:ICM,
author = Geurt Jongbloed,
title =The Iterative Convex Minorant Algorithm for
Nonparametric Estimation,
journal = j-J-COMPUT-GRAPH-STAT,
volume = 7,
number = 3,
pages =310--321,
month =sep,
year = 1998,
CODEN =,
ISSN = 1061-8600,
MRclass = 62G05,
MRnumber = 1 646 718,
bibdate = Sat Jan 2 17:33:21 MST 1999,
URL = http://www.amstat.org/publications/jcgs/abstracts98/jongbloed.html;,
acknowledgement = ack-nhfb,
}
--
I know about Roger Koenker and Pin Ng's very interesting nprq
package (and its earlier version cobs) where one can fit splines
with convexity constraints.
At the moment however, we are specifically interested in the ICM
algorithm for convex nonparametric regression.
Thank you for pointers in advance.
Martin Maechler, ETH Zurich
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] duplicate row importing issue
Or just read.table(..., header=FALSE, skip=1) Andy From: F Duan How about you delete the first row (column names) and then use header=FALSE in read.table()? Frank -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Herman, David (NIH/NIMH) Sent: Wednesday, July 14, 2004 11:13 To: '[EMAIL PROTECTED]' Subject: [R] duplicate row importing issue Hello, I'm simply trying to import a .txt file that has a column of about 30,000 pts. I found the file, but I'm getting an error: m - read.table(choose.files()) Error in row.names-.data.frame(`*tmp*`, value = row.names) : duplicate row.names are not allowed Any help with getting around this? I really appreciate all the help. Thanks dave [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Permutations
Dear Adaikalavan,
Now I've send a similar e-mail to Robin and Rolf, therefore I can deduce
that my first e-mail was not enough clear (sorry).
Therefore, I'm going to try it again with an explanation of why I don't
want those permutations:
Ive 12 elements in blocks of 3 elements and I want only to make
permutations inter-blocks (no intra-blocks) (sorry if the terminology is not
accurate), something similar to:
1 2 3 | 4 5 6 | 7 8 9 | 10 11 12 YES---1st permutation
1 3 2 | 4 5 6 | 7 8 9 | 10 11 12 NO because it's an intra-block
permutation of permutation 1
- -
3 2 1 | 4 5 6 | 7 8 9 | 10 11 12 NO because it's an intra-block
permutation of permutation 1
- - -
1 2 4 | 3 5 6 | 7 8 9 | 10 11 12 YES-2nd permutation
4 5 6 | 1 2 3 | 7 8 9 | 10 11 12 YES-3rd permutation
4 5 6 | 2 1 3 | 7 8 9 | 10 11 12 NO because it's an intra-block
permutation of permutation 3
- -
10 1 7 | 4 8 7 | 5 6 12 | 3 2 9YES---Xth permutation
1 10 7 | 4 8 7 | 5 6 12 | 3 2 9NO because it's an
intra-blockpermutation of permutation X
- -
So, what is a not correct permutation is an intra-block permutation of a
permutation created before.
Again, thanks for your time and suggestions,
Jordi Altirriba
PhD student
Hospital Clinic - Barcelona - Spain
P.S. Probably (it's the way of how I'm testing the algorithms now with
Excel) [sorry if it's stupid], could be interesting something similar to:
1 2 3 | 4 5 6 | 7 8 9 | 10 11 12 --- 1*2*3=6 | 4*5*6=120 | 7*8*9=504 |
10*11*12=1320
1 3 2 | 4 5 6 | 7 8 9 | 10 11 12 --- 1*3*2=6 | 4*5*6=120 | 7*8*9=504 |
10*11*12=1320
4 5 6 | 1 2 3 | 7 8 9 | 10 11 12 --- 4*5*6=120 | 1*2*3=6 | 7*8*9=504 |
10*11*12=1320
Results:
permutation1: 6 | 120 | 504 | 1320
permutation2: 6 | 120 | 504 | 1320
permutation3: 120 | 6 | 504 | 1320
Order the permutations according first to the first parameter, second to the
second...and to the fourth.
In this case it's the same:
permutation1: 6 | 120 | 504 | 1320
permutation2: 6 | 120 | 504 | 1320
permutation3: 120 | 6 | 504 | 1320
Therefore, if the first, second, third and fourth parameter of a
permutations have the same value that the next permutation it's because
there is an intra-block permutation. So, permutation 2 is an intra-block
permutation of permutation 1.
P.S. Sorry for having forgotten the title of the last e-mail
From: Adaikalavan Ramasamy [EMAIL PROTECTED]
To: Jordi Altirriba Gutiérrez [EMAIL PROTECTED]
CC: [EMAIL PROTECTED], R-help [EMAIL PROTECTED]
Subject: Re: [R] Permutations
Date: Wed, 14 Jul 2004 18:00:49 +0100
I think the issue here is in the two keywords - permutations or sample.
AFAIK, permutations should return all admissible (by some rule)
combinations. If this is a large number, as some have pointed out, then
one essentially takes a _sample_ of all admissible combinations. Since
you earlier mentioned that you only want 5-10 outputs, perhaps the
correct term is sampling with restrictions.
There main problem with Robin's method in that all elements within a row
are mutually exclusive to the other. e.g. only one of either 1, 4, 7, 10
can appear in block 1. Furthermore they can only appear in the first
slot of the first block (so no intra-block randomness). This limits the
number of possible outputs.
Can you clearly define the rules (with examples) for an admissible
combination ? They seem to have a different meaning every time I read
the mail. Maybe I am just confused.
On Wed, 2004-07-14 at 17:16, Jordi Altirriba Gutiérrez wrote:
Dear R users,
First of all, thanks to Rolf, Brad, Robin, Erich, Fernando and
Adaikalavan
for your time and suggestions.
Ive been testing some algorithms (sorry for the delay, Im very slow,
and
Im a completely beginner in Rs world).
First, the Robin algorithm.
I think that there is a problem because Ive done 200 permutations and
Ive found that these permutations are the same:
52 and 91, 99 and 110, 121 and 122, 51 and 141, 130 and 134.
Thanks again,
Jordi Altirriba
Hospital Clinic Barcelona - Spain
x - c(1,2,3,4,5,6,7,8,9,10,11,12)
dim(x) - c(3,4) a-matrix(1,200,12)
for (i in 1:200)
+ {
+ jj - t(apply(x,1,sample))
+ a[i,]-as.vector(jj)
+ }
a
[,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10] [,11] [,12]
[1,]7231 11648910 512
[2,]1297 11648 1210 5 3
[3,]7291 11345610 812
[4,] 108645 12729 111 3
[5,] 102 121 119786 4 5 3
[6,]7861594 11 1210 2 3
[7,]15 127264891011 3
[8,]159 1086423 71112
[9,]1 11672 1245910 8 3
[10,]45 12 10 119186 7 2 3
Re: [R] Permutations
On Wed, 2004-07-14 at 08:06, Rolf Turner wrote:
In respect of generating random ``restricted'' permutations, it
occurred to me as I was driving home last night If one is going
to invoke some kind of ``try again if it doesn't work procedure''
then one might as well keep it simple: Essentially use the rejection
method. Just generate a random permutation, and then check whether
it meets the restriction criterion. If yes, return that
permutation, if not, throw it away and try again.
This will definitely (???) genererate a genuinely random restricted
permutation. I figured that since a very large fraction of permutations
are acutally restricted permutions one wouldn't reject much of the
time, and so the rejection method should be pretty efficient.
I wrote the following code to do the work:
restr.perm2 - function () {
#
okay - function (x) {
m1 - cbind(1:12,rep(1:4,rep(3,4)))
m2 - m1[x,]
all((m2[,1] == m1[,1]) | (m2[,2] != m1[,2]))
}
#
repeat{
x - sample(12,12)
if(okay(x)) return(x)
}
}
I'm bothered again, but: I did a little test to see how many tries
it would take on average. On 1000 attempts I got a mean of 8.44
tries, with a standard deviation of 7.7610 (standard error of the
mean = 0.2454). The value of 7.76 is roughly consistent with
sqrt(1-p.hat)/p.hat = 7.92 that one gets from the geometric
distribution.
This would indicate that the fraction of ``restricted'' permutations
is something like p.hat = 1/8.44 = 0.1184834. Which is a lot less
than the rough estimate of (4.7 x 10^6)/12! approx. = 0.9853 from
Robert Baskin's back-of-the-envelope calculations.
What's going on/wrong?
Rolf,
I have not quite figured out what the issues are, but I took your
approach above and combined it with Gabor's f1a() function that was the
result of the recent thread on matching/counting rows between matrices
and came up with the following. The basic concept is that we know (or
think that we know) what the non-allowable set of intra-block
permutations are:
# Create non-allowable 'intra-block' permutations
# Need a generalizable way to do this, but
# good enough for now
a - permutations(3, 3, 1:3)
b - permutations(3, 3, 4:6)
c - permutations(3, 3, 7:9)
d - permutations(3, 3, 10:12)
intra - rbind(a[-1, ], b[-1, ], c[-1, ], d[-1, ])
'intra' looks like:
intra
[,1] [,2] [,3]
[1,]132
[2,]213
[3,]231
[4,]312
[5,]321
[6,]465
[7,]546
[8,]564
[9,]645
[10,]654
[11,]798
[12,]879
[13,]897
[14,]978
[15,]987
[16,] 10 12 11
[17,] 11 10 12
[18,] 11 12 10
[19,] 12 10 11
[20,] 12 11 10
With that in place, we can then take the randomly generated 'x', coerce
it into a 3 column matrix by row and use Gabor's function to check for
any matches of the rows of 3 in 'x' with the non-allowable permutations
in 'intra'.
Thus, the function will assign 'x' to the appropriate row in 'results'
(which is pre-allocated based upon the number of runs) if 'x' passes or
sets the row to NA's if it does not.
I then use complete.cases() to return only the valid rows. Presumably,
some of the randomly generated 'x' vectors could be duplicates, so I
also use unique():
restr.perm3 - function(runs)
{
results - matrix(numeric(runs * 12), ncol = 12)
# use Gabor's function to check for row matches
# between 'x' and 'intra' to filter out in okay()
f1a - function(a,b,sep=:)
{
f - function(...) paste(..., sep=sep)
a2 - do.call(f, as.data.frame(a))
b2 - do.call(f, as.data.frame(b))
c(table(c(b2,unique(a2)))[a2] - 1)
}
okay - function (x)
{
x.match - matrix(x, ncol = 3, byrow = TRUE)
all(f1a(x.match, intra) == 0)
}
for (i in 1:runs)
{
x - sample(12,12)
if (okay(x))
results[i, ] - x
else
results[i, ] - rep(NA, 12)
}
unique(results[complete.cases(results), ])
}
So, with all that in place, we can then do something like the following:
r - replicate(10, restr.perm3(1000))
str(r)
List of 10
$ : num [1:934, 1:12] 10 6 8 7 7 11 1 1 8 4 ...
$ : num [1:944, 1:12] 7 4 4 11 1 11 8 3 3 9 ...
$ : num [1:953, 1:12] 8 4 11 1 11 6 7 11 9 10 ...
$ : num [1:951, 1:12] 1 2 1 4 4 10 8 10 3 12 ...
$ : num [1:949, 1:12] 1 7 11 9 2 2 11 7 11 4 ...
$ : num [1:937, 1:12] 3 3 11 10 4 8 12 10 3 2 ...
$ : num [1:952, 1:12] 7 3 1 6 4 4 4 11 2 8 ...
$ : num [1:946, 1:12] 1 9 3 10 11 6 1 7 8 11 ...
$ : num [1:945, 1:12] 8 9 1 2 8 3 4 1 7 11 ...
$ : num [1:933, 1:12] 9 1 12 3 4 1 10 1 1 8 ...
So the above would suggest that indeed, the restricted permutations are
a fairly high proportion of the total.
I went ahead and did the following 50 replicates of 1000 each:
system.time(r - replicate(50, restr.perm3(1000)))
[1] 91.20 0.06 92.98 0.00 0.00
mean(unlist(lapply(r, nrow)))
[R] simultaneous confidence intervals --- question
Dear all, is there a R function to construct simultaneous confidence intervals for multinomial proportions? Best wishes jl __ Acabe com aquelas janelinhas que pulam na sua tela. AntiPop-up UOL - É grátis! http://antipopup.uol.com.br/ __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] PCA in R
Hello, I'm attempting to run a PCA on an example data set. I ran it just fine, but I don't know how to few the output? I listed what the variable got stored in it, but I don't know how I can get anything else out of it. Are there other ways to view the results? Also, I'm confused about the last line 6 variables and 8 observations Aren't the rows the variables and the columns the observations? (NOTE: if anyone knows a good guide for doing a PCA on an example data set, from start to finish, it would be greatly appreciated) m [,1] [,2] [,3] [,4] [,5] [,6] [1,]123456 [2,]343781 [3,]123456 [4,]874143 [5,]123456 [6,]458213 [7,]987678 [8,]137350 pc.cr-princomp(m,cor=TRUE) pc.cr Call: princomp(x = m, cor = TRUE) Standard deviations: Comp.1 Comp.2 Comp.3 Comp.4 Comp.5 Comp.6 1.545609e+00 1.407093e+00 9.886649e-01 7.539927e-01 2.919276e-01 2.460515e-09 6 variables and 8 observations. thanks! Dave [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] PCA in R
On 07/14/04 16:05, Herman, David (NIH/NIMH) wrote: Hello, I'm attempting to run a PCA on an example data set. I ran it just fine, but I don't know how to few the output? Take a look at the help file for prcomp, especially the bottom of it. (This is completely general advice for any R function.) You will see under See also a number of things listed, and if you look them up or try them, then you will see several different ways of viewing the result. Under that is Examples, which provides more hints still. And, if this isn't enough, type example(prcomp) on the command line. The biplot function is especially nice. Jon -- Jonathan Baron, Professor of Psychology, University of Pennsylvania Home page:http://www.sas.upenn.edu/~baron R search page:http://finzi.psych.upenn.edu/ __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] PCA in R
When all else fails, RTFM; e.g., see ?princomp and read it in its entirety. Then maybe also try running example(princomp). Andy From: Herman, David (NIH/NIMH) Hello, I'm attempting to run a PCA on an example data set. I ran it just fine, but I don't know how to few the output? I listed what the variable got stored in it, but I don't know how I can get anything else out of it. Are there other ways to view the results? Also, I'm confused about the last line 6 variables and 8 observations Aren't the rows the variables and the columns the observations? (NOTE: if anyone knows a good guide for doing a PCA on an example data set, from start to finish, it would be greatly appreciated) m [,1] [,2] [,3] [,4] [,5] [,6] [1,]123456 [2,]343781 [3,]123456 [4,]874143 [5,]123456 [6,]458213 [7,]987678 [8,]137350 pc.cr-princomp(m,cor=TRUE) pc.cr Call: princomp(x = m, cor = TRUE) Standard deviations: Comp.1 Comp.2 Comp.3 Comp.4 Comp.5 Comp.6 1.545609e+00 1.407093e+00 9.886649e-01 7.539927e-01 2.919276e-01 2.460515e-09 6 variables and 8 observations. thanks! Dave [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Permutations
Based on your description below and our off-list
discussion I gather than the problem is equivalent
to sampling ordered permutations in the sense
defined by Erich (i.e. permutations which are
increasing within blocks) WITHOUT replacement.
Actually one of your valid permutations in your
example is not ordered but it seems arbitrary
which representative of the 3!^4 intrablock
permutations is used which is why I believe this
solves your problem.
The following code (not extensively tested) does this:
ordered.perm - function(N) {
samp - function() c(apply(matrix(sample(12,12),3),2,sort))
z - vector(length=N, mode=character)
for(i in 1:N)
while( (z[i]-paste(samp(),collapse= )) %in% z[seq(len=i-1)] ) {}
matrix(as.numeric(unlist(strsplit(z, split = ))), nc = 12, byrow = TRUE)
}
ordered.perm(3) # test run
In the above, samp(), which is from my previous
response, gets one sample WITH replacement. It generates
an unrestricted permutation of 12 and then projects that
onto its associated ordered permutation of 12. Note that
samp does not use rejection. That comes later which is
why there are relatively few rejections.
Each iteration of the for loop gets one sample without
replacement storing the unrejected sammples as
character strings in vector z. Each iteration of this
for loop uses a while to call samp() repeatedly until it
finds a sample that has not previously been generated
(i.e. rejecting the others). The line beginning
with matrix converts the strings to numbers.
Jordi Altirriba Gutirrez altirriba at hotmail.com writes:
:
: Dear Adaikalavan,
: Now I've send a similar e-mail to Robin and Rolf, therefore I can deduce
: that my first e-mail was not enough clear (sorry).
: Therefore, I'm going to try it again with an explanation of why I don't
: want those permutations:
:
: Ive 12 elements in blocks of 3 elements and I want only to make
: permutations inter-blocks (no intra-blocks) (sorry if the terminology is not
: accurate), something similar to:
:
: 1 2 3 | 4 5 6 | 7 8 9 | 10 11 12 YES---1st permutation
:
: 1 3 2 | 4 5 6 | 7 8 9 | 10 11 12 NO because it's an intra-block
: permutation of permutation 1
:- -
: 3 2 1 | 4 5 6 | 7 8 9 | 10 11 12 NO because it's an intra-block
: permutation of permutation 1
: - - -
: 1 2 4 | 3 5 6 | 7 8 9 | 10 11 12 YES-2nd permutation
:
: 4 5 6 | 1 2 3 | 7 8 9 | 10 11 12 YES-3rd permutation
:
: 4 5 6 | 2 1 3 | 7 8 9 | 10 11 12 NO because it's an intra-block
: permutation of permutation 3
:- -
: 10 1 7 | 4 8 7 | 5 6 12 | 3 2 9YES---Xth permutation
:
: 1 10 7 | 4 8 7 | 5 6 12 | 3 2 9NO because it's an
: intra-blockpermutation of permutation X
: - -
:
: So, what is a not correct permutation is an intra-block permutation of a
: permutation created before.
:
: Again, thanks for your time and suggestions,
:
: Jordi Altirriba
: PhD student
: Hospital Clinic - Barcelona - Spain
:
: P.S. Probably (it's the way of how I'm testing the algorithms now with
: Excel) [sorry if it's stupid], could be interesting something similar to:
:
: 1 2 3 | 4 5 6 | 7 8 9 | 10 11 12 --- 1*2*3=6 | 4*5*6=120 | 7*8*9=504 |
: 10*11*12=1320
:
: 1 3 2 | 4 5 6 | 7 8 9 | 10 11 12 --- 1*3*2=6 | 4*5*6=120 | 7*8*9=504 |
: 10*11*12=1320
:
: 4 5 6 | 1 2 3 | 7 8 9 | 10 11 12 --- 4*5*6=120 | 1*2*3=6 | 7*8*9=504 |
: 10*11*12=1320
:
:
: Results:
: permutation1: 6 | 120 | 504 | 1320
: permutation2: 6 | 120 | 504 | 1320
: permutation3: 120 | 6 | 504 | 1320
:
: Order the permutations according first to the first parameter, second to the
: second...and to the fourth.
:
: In this case it's the same:
: permutation1: 6 | 120 | 504 | 1320
: permutation2: 6 | 120 | 504 | 1320
: permutation3: 120 | 6 | 504 | 1320
:
: Therefore, if the first, second, third and fourth parameter of a
: permutations have the same value that the next permutation it's because
: there is an intra-block permutation. So, permutation 2 is an intra-block
: permutation of permutation 1.
:
: P.S. Sorry for having forgotten the title of the last e-mail
:
:
: From: Adaikalavan Ramasamy ramasamy at cancer.org.uk
: To: Jordi Altirriba Gutirrez altirriba at hotmail.com
: CC: rksh at soc.soton.ac.uk, R-help r-help at stat.math.ethz.ch
: Subject: Re: [R] Permutations
: Date: Wed, 14 Jul 2004 18:00:49 +0100
:
: I think the issue here is in the two keywords - permutations or sample.
:
: AFAIK, permutations should return all admissible (by some rule)
: combinations. If this is a large number, as some have pointed out, then
: one essentially takes a _sample_ of all admissible combinations. Since
: you earlier mentioned that you only want 5-10 outputs, perhaps the
: correct term is sampling with restrictions.
:
: There main problem with Robin's method in that all elements within a row
: are mutually exclusive to the other. e.g. only one of either 1, 4, 7, 10
: can appear in block 1. Furthermore they can only appear in the first
[R] using mean in by(x,y,mean)
Dear list friends I fail to understand how to find means for multiple groups using the by() function. Help would be appreciated. Thanks. Bill x - runif(20,0,10) group - rep(c(A,B),10) df -data.frame(x,group) #df#show the data rm(x,group) attach(df) sd(x) # sd is defined mean(x) #so is mean by(x,group,sd) #this works for both groups by(x,group,mean) #this does not produces this output: x - runif(20,0,10) group - rep(c(A,B),10) df -data.frame(x,group) #df#show the data rm(x,group) attach(df) sd(x) # sd is defined [1] 2.952699 mean(x) #so is mean [1] 5.026441 by(x,group,sd) #this works for both groups INDICES: A [1] 2.813504 INDICES: B [1] 3.236663 by(x,group,mean) #this does not Error in FUN(X[[as.integer(1)]], ...) : couldn't find function FUN I am using a Mac OS 10.3.4 running R 1.9.1 -- William Revelle http://pmc.psych.northwestern.edu/revelle.html Department of Psychology, Northwestern University Personality Project: http://personality-project.org/personality.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] using mean in by(x,y,mean)
William Revelle wrote: Dear list friends I fail to understand how to find means for multiple groups using the by() function. Help would be appreciated. Thanks. Bill x - runif(20,0,10) group - rep(c(A,B),10) df -data.frame(x,group) #df#show the data rm(x,group) attach(df) sd(x) # sd is defined mean(x) #so is mean by(x,group,sd) #this works for both groups by(x,group,mean) #this does not produces this output: x - runif(20,0,10) group - rep(c(A,B),10) df -data.frame(x,group) #df#show the data rm(x,group) attach(df) sd(x) # sd is defined [1] 2.952699 mean(x) #so is mean [1] 5.026441 by(x,group,sd) #this works for both groups INDICES: A [1] 2.813504 INDICES: B [1] 3.236663 by(x,group,mean) #this does not Error in FUN(X[[as.integer(1)]], ...) : couldn't find function FUN I am using a Mac OS 10.3.4 running R 1.9.1 Hi Bill, Works for me. Do you have a local mean defined? Try rm(mean) and re-try. R.version _ platform i386-pc-mingw32 arch i386 os mingw32 system i386, mingw32 status major1 minor9.1 year 2004 month06 day 21 language R --sundar __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] using mean in by(x,y,mean)
From: Sundar Dorai-Raj William Revelle wrote: Dear list friends I fail to understand how to find means for multiple groups using the by() function. Help would be appreciated. Thanks. Bill x - runif(20,0,10) group - rep(c(A,B),10) df -data.frame(x,group) #df#show the data rm(x,group) attach(df) sd(x) # sd is defined mean(x) #so is mean by(x,group,sd) #this works for both groups by(x,group,mean) #this does not produces this output: x - runif(20,0,10) group - rep(c(A,B),10) df -data.frame(x,group) #df#show the data rm(x,group) attach(df) sd(x) # sd is defined [1] 2.952699 mean(x) #so is mean [1] 5.026441 by(x,group,sd) #this works for both groups INDICES: A [1] 2.813504 -- -- INDICES: B [1] 3.236663 by(x,group,mean) #this does not Error in FUN(X[[as.integer(1)]], ...) : couldn't find function FUN I am using a Mac OS 10.3.4 running R 1.9.1 Hi Bill, Works for me. Do you have a local mean defined? Try rm(mean) and re-try. R.version _ platform i386-pc-mingw32 arch i386 os mingw32 system i386, mingw32 status major1 minor9.1 year 2004 month06 day 21 language R --sundar Or try replacing `mean' with `base:::mean'. BTW, why not just use tapply(x, group, mean)? Andy __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] MASS package?
Did the MASS package disappear? Specifically, I'm looking for a function to find the MCD (robust measure of shape and location) for a multi-dimensional data matrix. Anyone know anything about this? Thanks, Jo Jo Hardin Assistant Professor Department of Mathematics Pomona College 610 N. College Ave. Claremont, CA 91711 909-607-8717 [EMAIL PROTECTED] [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] RODBC repeated rows problem
Hello everyone! I'm having the dreaded repeated rows problem in RODBC. Specifically, when I have a NULL value in a column, odbcFetchRows reads the value not as NULL or NA but as the most recent non-NULL value in the column. If there is no such non-NULL column, odbcFetchRows reads the value as 0. Let me give you an example (though you won't be able to run it with the product table--just use some table you've defined): head(sqlQuery(channel, select NULL from product)) c(0, 0, 0, 0, 0, 0) 1 0 2 0 3 0 4 0 5 0 6 0 Obviously, I'd like to see NAs here, not 0s. Let me give you some of my specs: version _ platform x86_64-unknown-linux-gnu arch x86_64 os linux-gnu system x86_64, linux-gnu status major1 minor9.1 year 2004 month06 day 21 language R Other specs: I'm reading a unixODBC ODBC driver. When I run from isql and try the same query, I get, as I might expect, a bunch of blanks. SQL select top 5 NULL from product ++ || ++ || || || || || ++ SQLRowCount returns 5 5 rows fetched It shouldn't be relevant, but the ODBC is powered by a FreeTDS driver hooked to an MS SQL Server 2000 Professional. I noticed a previous note from Professor Ripley reporting that this problem had been fixed with a version of RODBC from long ago. I am running RODBC version 1.04, the latest version, however, so I'm at a loss as to what to do. One last note: I looked at the relevant line from sql.h that defines my SQL_NULL_DATA, which RODBC (presumably) compares to its results. It says: #define SQL_NULL_DATA (-1) Unfortunately, I have no idea what this should mean to me :( Are there any further diagnostics I can run? Any new versions of software I should install? Any workarounds for the time being? Thanks for any help you can provide. Kevin __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] MASS package?
Search on the R site ought to show you. cov.rob() is in MASS (part of the `VR' bundle, and should be part of all R distribution). There's also the rrcov package with the covMcd() function. Andy From: Johanna Hardin Did the MASS package disappear? Specifically, I'm looking for a function to find the MCD (robust measure of shape and location) for a multi-dimensional data matrix. Anyone know anything about this? Thanks, Jo Jo Hardin Assistant Professor Department of Mathematics Pomona College 610 N. College Ave. Claremont, CA 91711 909-607-8717 [EMAIL PROTECTED] [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] MASS package?
On Wed, 2004-07-14 at 17:05, Johanna Hardin wrote: Did the MASS package disappear? Specifically, I'm looking for a function to find the MCD (robust measure of shape and location) for a multi-dimensional data matrix. Anyone know anything about this? Try: library(MASS) ?cov.rob It's there, unless you have a corrupted/incomplete installation. HTH, Marc Schwartz __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Permutations
Just in case my last post is wrong in assuming that
ordered repesentatives of the 3!^4 intragroup permutations
are adequate, the following minor variation of the routine
in my last post will provide an unordered representative.
Its actually easy to do since sample(12,12), which we
already calculated, is that representative so we merely
store it.
ordered.perm2 - function(N) {
samp - function() c(apply(matrix(z[i,] - sample(12,12),3),2,sort))
s - vector(length = N, mode = character)
z - matrix(nr = N, nc = 12)
for(i in 1:N)
while( (s[i]-paste(samp(),collapse= )) %in% s[seq(len=i-1)] ) {}
z
}
# test
set.seed(1)
ordered.perm(3)
Gabor Grothendieck ggrothendieck at myway.com writes:
:
: Based on your description below and our off-list
: discussion I gather than the problem is equivalent
: to sampling ordered permutations in the sense
: defined by Erich (i.e. permutations which are
: increasing within blocks) WITHOUT replacement.
:
: Actually one of your valid permutations in your
: example is not ordered but it seems arbitrary
: which representative of the 3!^4 intrablock
: permutations is used which is why I believe this
: solves your problem.
:
: The following code (not extensively tested) does this:
:
: ordered.perm - function(N) {
:samp - function() c(apply(matrix(sample(12,12),3),2,sort))
:z - vector(length=N, mode=character)
:for(i in 1:N)
: while( (z[i]-paste(samp(),collapse= )) %in% z[seq(len=i-1)] ) {}
:matrix(as.numeric(unlist(strsplit(z, split = ))), nc = 12, byrow =
TRUE)
: }
: ordered.perm(3) # test run
:
: In the above, samp(), which is from my previous
: response, gets one sample WITH replacement. It generates
: an unrestricted permutation of 12 and then projects that
: onto its associated ordered permutation of 12. Note that
: samp does not use rejection. That comes later which is
: why there are relatively few rejections.
:
: Each iteration of the for loop gets one sample without
: replacement storing the unrejected sammples as
: character strings in vector z. Each iteration of this
: for loop uses a while to call samp() repeatedly until it
: finds a sample that has not previously been generated
: (i.e. rejecting the others). The line beginning
: with matrix converts the strings to numbers.
:
: Jordi Altirriba Gutirrez altirriba at hotmail.com writes:
:
: :
: : Dear Adaikalavan,
: : Now I've send a similar e-mail to Robin and Rolf, therefore I can deduce
: : that my first e-mail was not enough clear (sorry).
: : Therefore, I'm going to try it again with an explanation of why I don't
: : want those permutations:
: :
: : Ive 12 elements in blocks of 3 elements and I want only to make
: : permutations inter-blocks (no intra-blocks) (sorry if the terminology is
not
: : accurate), something similar to:
: :
: : 1 2 3 | 4 5 6 | 7 8 9 | 10 11 12 YES---1st permutation
: :
: : 1 3 2 | 4 5 6 | 7 8 9 | 10 11 12 NO because it's an intra-block
: : permutation of permutation 1
: :- -
: : 3 2 1 | 4 5 6 | 7 8 9 | 10 11 12 NO because it's an intra-block
: : permutation of permutation 1
: : - - -
: : 1 2 4 | 3 5 6 | 7 8 9 | 10 11 12 YES-2nd permutation
: :
: : 4 5 6 | 1 2 3 | 7 8 9 | 10 11 12 YES-3rd permutation
: :
: : 4 5 6 | 2 1 3 | 7 8 9 | 10 11 12 NO because it's an intra-block
: : permutation of permutation 3
: :- -
: : 10 1 7 | 4 8 7 | 5 6 12 | 3 2 9YES---Xth permutation
: :
: : 1 10 7 | 4 8 7 | 5 6 12 | 3 2 9NO because it's an
: : intra-blockpermutation of permutation X
: : - -
: :
: : So, what is a not correct permutation is an intra-block permutation of
a
: : permutation created before.
: :
: : Again, thanks for your time and suggestions,
: :
: : Jordi Altirriba
: : PhD student
: : Hospital Clinic - Barcelona - Spain
: :
: : P.S. Probably (it's the way of how I'm testing the algorithms now with
: : Excel) [sorry if it's stupid], could be interesting something similar to:
: :
: : 1 2 3 | 4 5 6 | 7 8 9 | 10 11 12 --- 1*2*3=6 | 4*5*6=120 | 7*8*9=504 |
: : 10*11*12=1320
: :
: : 1 3 2 | 4 5 6 | 7 8 9 | 10 11 12 --- 1*3*2=6 | 4*5*6=120 | 7*8*9=504 |
: : 10*11*12=1320
: :
: : 4 5 6 | 1 2 3 | 7 8 9 | 10 11 12 --- 4*5*6=120 | 1*2*3=6 | 7*8*9=504 |
: : 10*11*12=1320
: :
: :
: : Results:
: : permutation1: 6 | 120 | 504 | 1320
: : permutation2: 6 | 120 | 504 | 1320
: : permutation3: 120 | 6 | 504 | 1320
: :
: : Order the permutations according first to the first parameter, second to
the
: : second...and to the fourth.
: :
: : In this case it's the same:
: : permutation1: 6 | 120 | 504 | 1320
: : permutation2: 6 | 120 | 504 | 1320
: : permutation3: 120 | 6 | 504 | 1320
: :
: : Therefore, if the first, second, third and fourth parameter of a
: : permutations have the same value that the next permutation it's because
: : there is an intra-block permutation. So, permutation 2 is an intra-
block
: : permutation of permutation 1.
: :
:
[R] questions about R
Hi, I try to compute critical value for multivariate normal distribution, and I find the crit(fit, const = c(0, 1), d = 1, cov = 0.95, rdf = 0) which seems to compute critical value. However, I can't compute right critical value for multivariate normal distribution which I want. I don't know how to solve my question. I send my question in the file. Would you help me to solve it, please? My English is not very well, but I appreciate you very much. Chia-Yi - ,! Yahoo!Messenger6.0__ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] older versions of R (v1.0)
Hi folks, Does anyone know where I can find downloads for old versions of R, around version 1.0? I found the downloads for unix but not windows. Any help would be greatly appreciated! Thank you, Kathryn __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
