[R] Updating packages to R 2.1.0
A very handy new feature of 2.1.0 is that update.packages(ask=FALSE, checkBuilt=TRUE) will re-install all your packages with current versions built under 2.1.0. This works for both Unix-alikes and Windows (and for MacOS in due course). If you have many packages installed, allow some time (maybe an hour for all the CRAN packages on a P4 2.4GHz Linux box). -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] dynamic function loading
Omar Lakkis wrote: How can I do dynamic function loading in R? I have a list of R functions in the database and want to, dynamicly, execute teh function that corresponds to my in comming data. Is there a way to do that without a big if/else? I don't know whether I understand correctly, but I guess you are loooking for object oriented approaches. So write a generic function and its methods that correspond to your data objects. You might want to read the green book on S4 classes. Uwe Ligges __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] error due to locales again
Yes. Please do beta-test *before* release: R 2.1.0 is released.
If no one actually tests R in your language, we will not discover that
there are problems in that language. Over the the users to beta-test when
asked to.
On Tue, 19 Apr 2005, ronggui wrote:
example(lm)
Error in switch(x[2], "1250" = return("ISO 8859-2"), "1251" = return("KOI8-U"),
:
argument is missing, with no default
Sys.getlocale(category = "LC_ALL")
[1] "LC_COLLATE=Chinese_People's Republic of China.936;LC_CTYPE=Chinese_People's
Republic of China.936;LC_MONETARY=Chinese_People's Republic of
China.936;LC_NUMERIC=C;LC_TIME=Chinese_People's Republic of China.936"
version
_
platform i386-pc-mingw32
arch i386
os mingw32
system i386, mingw32
status beta
major2
minor1.0
year 2005
month04
day 16
language R
--
Brian D. Ripley, [EMAIL PROTECTED]
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UKFax: +44 1865 272595
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] error due to locales again
>example(lm)
Error in switch(x[2], "1250" = return("ISO 8859-2"), "1251" = return("KOI8-U"),
:
argument is missing, with no default
> Sys.getlocale(category = "LC_ALL")
[1] "LC_COLLATE=Chinese_People's Republic of
China.936;LC_CTYPE=Chinese_People's Republic of
China.936;LC_MONETARY=Chinese_People's Republic of
China.936;LC_NUMERIC=C;LC_TIME=Chinese_People's Republic of China.936"
> version
_
platform i386-pc-mingw32
arch i386
os mingw32
system i386, mingw32
status beta
major2
minor1.0
year 2005
month04
day 16
language R
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] using imported tables
I guess one of the reasons that you have not had a reply is that you have not
followed the posting guide. If you give the list something to work with (a
small reproducible example)
You use the word package which in R is very precise, but which I think you are
using to describe the file you are reading. This amkes it harder for us to
understand what your problem is.
if you type ?read.table and read the help you will find that the parameter
col.names is a "a vector of optional names for the variables. The default is to
use '"V"' followed by the column number." Typically this would look something
like col.names = c("ColA","ColB", ..., "ColZ") That is one for each column in
the matrix or data.frame.
But I'm guessing. If I saw the code I might see something else that you've done
to to see what you mean by "assigns a name to the entire table."
Now's a good time to reread the posting guide, assuming that you have already
done so.
Tom
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] Behalf Of Owen Buchner
> Sent: Tuesday, 19 April 2005 10:16 AM
> To: R-help@stat.math.ethz.ch
> Subject: [R] using imported tables
>
>
> I've recently found out using read.table i can insert a table
> into R from a .dat
> file. The problem i'm having is now that i've got R to read
> the package i want
> to use the data from each column. I've tried adding titles
> to each column
> using col.name, but it assigns a name to the entire table and
> wont recognize
> the name i gave the table later in programming. Is there
> something i'm doing
> wrong or am i missing an important step? I'd appreciate any help.
>
> __
> R-help@stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
> http://www.R-project.org/posting-guide.html
>
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Compatability with Tiger OS X 10.4
No one _can_ know. Mac OS 10.4 is unreleased, and things can easily break before release. It is working on pre-releases. One problem is that Mac OS 10.4 (don't think you need two tens) will use gcc 4.0.0, which is also unreleased and its release candidates have problems compiling R (see the R-admin manual for R 2.1.0), including on Mac OS 10.4: there are also questions about how `properly' it runs R. The biggest problem we found in gcc 4.0.0 has been fixed within the last week, and it is unclear if this fix will make Mac OS 10.4. The R Mac OS X maintainers are working hard (thank you Stefano and Simon) on this and I expect them to succeed, but quite possibly by using patched tools or non-standard options (such as using a static Fortran runtime). Watch this space (or R-sig-mac if it needs to be technical). On Mon, 18 Apr 2005, Hemant Ishwaran wrote: Does anyone know if R will install and run properly on the soon to be released Tiger operating system (Mac OS X 10.4)? There must be several Mac users out there (like myself) who will be considering the upgrade. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Odd diagnostic plots in mixed-effects models
Dear R community, In the excellent nlme package the default diagnostic plot graphs the innermost residuals against innermost fitted values. I recently fit a mixed-effects model in which there was a very clear positive linear trend in this plot. I inferred that this trend occurred because my fixed effect was a two-level factor, and my random effect was a 12-level factor. The negative residuals were associated with negative random effects (because of shrinkage, I assume), and the positive with positive. The fixed effects explained little varaition. Therefore plotting the innermost residuals against the innermost fitted values had the negative residuals to the left and the positive residuals to the right, occasioning a trend. My questions are: is it (as I suspect) harmless, or does it suggest that the model is lacking? And, is this effect likely to compromise the interpretation of any of the other standard diagnostic plots (eg qqnorm)? Thanks much for any thoughts, Andrew -- Andrew Robinson Ph: 208 885 7115 Department of Forest Resources Fa: 208 885 6226 University of Idaho E : [EMAIL PROTECTED] PO Box 441133W : http://www.uidaho.edu/~andrewr Moscow ID 83843 Or: http://www.biometrics.uidaho.edu No statement above necessarily represents my employer's opinion. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] using imported tables
I've recently found out using read.table i can insert a table into R from a .dat file. The problem i'm having is now that i've got R to read the package i want to use the data from each column. I've tried adding titles to each column using col.name, but it assigns a name to the entire table and wont recognize the name i gave the table later in programming. Is there something i'm doing wrong or am i missing an important step? I'd appreciate any help. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Compatability with Tiger OS X 10.4
don't know about the binaries, but I'd be surprised if you couldn't compile it. You've given me something to think about, though, as to whether I want to upgrade on my "analytic box". On 4/18/05, Hemant Ishwaran <[EMAIL PROTECTED]> wrote: > > Does anyone know if R will install and run properly on the > soon to be released Tiger operating system (Mac OS X 10.4)? > There must be several Mac users out there (like myself) who > will be considering the upgrade. > -- > Hemant Ishwaran > Staff, Dept of Quantitative Health Sciences, Cleveland Clinic Foundation > Adjunct Professor, Dept of Statistics, Case University > > http://www.bio.ri.ccf.org/Resume/Pages/Ishwaran/ishwaran.html > > __ > R-help@stat.math.ethz.ch mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! > http://www.R-project.org/posting-guide.html > -- --- David L. Van Brunt, Ph.D. mailto:[EMAIL PROTECTED] [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Listas de E-mails para Mala Direta , Marketing Direto
Cadastros para mala direta livre de spam programas grátis para e-mail marketing Cadastros para mala direta livre de spam e-mails para mala direta por classe social: http://www.gueb.de/divulgamail e-mails para mala direta por classe social Listas de E-mails para Mala Direta , Marketing Direto programas grátis para e-mail marketing cadastros de e-mails divididos por segmento Cadastros para mala direta livre de spam: http://www.gueb.de/divulgamail __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Compatability with Tiger OS X 10.4
Does anyone know if R will install and run properly on the soon to be released Tiger operating system (Mac OS X 10.4)? There must be several Mac users out there (like myself) who will be considering the upgrade. -- Hemant Ishwaran Staff, Dept of Quantitative Health Sciences, Cleveland Clinic Foundation Adjunct Professor, Dept of Statistics, Case University http://www.bio.ri.ccf.org/Resume/Pages/Ishwaran/ishwaran.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Barplot and colors for legend
Well I don't know how I can live with myself. I guess I can't wait for the site to mirror itself in case someone thinks I'm yesterday's man. ;-) > -Original Message- > From: [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED] Behalf Of Achim Zeileis > Sent: Monday, 18 April 2005 10:56 PM ... > Subject: Re: [R] Barplot and colors for legend ... > This must be an old version of R ;-) ... Tom Mulholland Perth, WA, Australia. ,-_|\ / \ ?_,-._/ v __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] longer object length, is not a multiple of shorter object length in: kappa * gcounts
I see. Thanks, Andy! Hui Liaw, Andy wrote: I suspect you set the bandwidth too small. Consider: x <- sort(runif(100)) * 100 min(diff(x)) [1] 0.001700642 library(KernSmooth) KernSmooth 2.22 installed Copyright M. P. Wand 1997 dx <- bkde(x, bandwidth=.001) Warning message: longer object length is not a multiple of shorter object length in: kappa * gcounts Andy From: Hui Han Hi, I was using a density estimation function as follows: > est <- KernSmooth::bkde(x3, bandwidth=10) When setting bandwidth less than 5, I got the error "longer object length, is not a multiple of shorter object length in: kappa * gcounts ". I wonder if there is anybody who can explain the error for me? Thanks! Hui __ R-help@stat.math.ethz.ch mailing list https://stat.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, contains information of Merck & Co., Inc. (One Merck Drive, Whitehouse Station, New Jersey, USA 08889), and/or its affiliates (which may be known outside the United States as Merck Frosst, Merck Sharp & Dohme or MSD and in Japan, as Banyu) that may be confidential, proprietary copyrighted and/or legally privileged. It is intended solely for the use of the individual or entity named on this message. If you are not the intended recipient, and have received this message in error, please notify us immediately by reply e-mail and then delete it from your system. -- __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] standard normal gaussian
?pnorm For more information, try "help.start()" -> "An Introduction to R" -> "Probability distributions". spencer graves simone gabbriellini wrote: Hello List, I need to calculate the p-value (in a standard normal gaussian) for a Z value I have. I guess if I had to calculate by myself the value of the area in my function or refer to some already made function in R. Any hints? Thank you very much, Simone __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] longer object length, is not a multiple of shorter object length in: kappa * gcounts
I suspect you set the bandwidth too small. Consider: > x <- sort(runif(100)) * 100 > min(diff(x)) [1] 0.001700642 > library(KernSmooth) KernSmooth 2.22 installed Copyright M. P. Wand 1997 > dx <- bkde(x, bandwidth=.001) Warning message: longer object length is not a multiple of shorter object length in: kappa * gcounts Andy > From: Hui Han > > Hi, > > I was using a density estimation function as follows: > > est <- KernSmooth::bkde(x3, bandwidth=10) > When setting bandwidth less than 5, I got the error "longer object > length, is not a multiple of shorter object length in: kappa > * gcounts ". > > I wonder if there is anybody who can explain the error for me? > > Thanks! > Hui > > __ > R-help@stat.math.ethz.ch mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! > http://www.R-project.org/posting-guide.html > > > __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Pearson corelation and p-value for matrix
> From: John Fox
>
> Dear Andy,
>
> Very nice! (My point was that if this is a one-time thing, for Dren to
> puzzle over it is probably more time-consuming than simply doing it
> inefficiently.)
There's a memory/speed tradeoff. The loop avoids creating multiple
6000x6000 matrices that my version would require, and one of those would
take up 274MB! I could not run my function on a 6000x6000 case on my laptop
(nor rcorr). The example I showed was 2000x2000. Also, as Frank pointed
out, it's not flexible. As John said, it's good for one-time use. For more
general consumption, one would want to write more flexible and robust code,
in which case one might very well re-invent rcorr...
Cheers,
Andy
> Regards,
> John
>
>
> John Fox
> Department of Sociology
> McMaster University
> Hamilton, Ontario
> Canada L8S 4M4
> 905-525-9140x23604
> http://socserv.mcmaster.ca/jfox
>
>
> > -Original Message-
> > From: Liaw, Andy [mailto:[EMAIL PROTECTED]
> > Sent: Monday, April 18, 2005 2:40 PM
> > To: 'John Fox'; 'Dren Scott'
> > Cc: 'R-Help'
> > Subject: RE: [R] Pearson corelation and p-value for matrix
> >
> > I believe this will do:
> >
> > cor.pval2 <- function(x, alternative="two-sided") {
> > corMat <- cor(x, use=if (any(is.na(x)))
> > "pairwise.complete.obs" else
> > "all")
> > df <- crossprod(!is.na(x)) - 2
> > STATISTIC <- sqrt(df) * corMat / sqrt(1 - corMat^2)
> > p <- pt(STATISTIC, df)
> > p <- if (alternative == "less") {
> > p
> > } else if (alternative == "greater") {
> > 1 - p
> > } else 2 * pmin(p, 1 - p)
> > p
> > }
> >
> > Some test:
> >
> > > set.seed(1)
> > > x <- matrix(runif(2e3 * 1e2), 2e3)
> > > system.time(res1 <- cor.pval(t(x)), gcFirst=TRUE)
> > [1] 17.28 0.77 18.16NANA
> > > system.time(res2 <- cor.pval2(t(x)), gcFirst=TRUE)
> > [1] 19.51 1.05 20.70NANA
> > > max(abs(res1 - res2))
> > [1] 0
> > > x[c(1, 3, 6), c(2, 4)] <- NA
> > > x[30, 61] <- NA
> > > system.time(res2 <- cor.pval2(t(x)), gcFirst=TRUE)
> > [1] 24.48 0.71 25.28NANA
> >
> > This is a bit slower because of the extra computation for
> > "df". One can try to save some computation by only computing
> > with the lower (or upper) triangular part.
> >
> > Cheers,
> > Andy
> >
> > > From: John Fox
> > >
> > > Dear Dren,
> > >
> > > Since cor(), on which Andy's solution is based, can compute
> > > pairwise-present correlations, you could adapt his function
> > -- you'll
> > > have to adjust the df for each pair. Alternatively, you
> > could probably
> > > save some time (programming time + computer time) by just
> using my
> > > solution:
> > >
> > > > R <- diag(100)
> > > > R[upper.tri(R)] <- R[lower.tri(R)] <- .5
> > > > library(mvtnorm)
> > > > X <- rmvnorm(6000, sigma=R)
> > > > system.time(for (i in 1:50) cor.pvalues(X), gc=TRUE)
> > > [1] 518.19 1.11 520.23 NA NA
> > >
> > > I know that time is money, but nine minutes (on my machine)
> > probably
> > > won't bankrupt anyone.
> > >
> > > Regards,
> > > John
> > >
> > >
> > > John Fox
> > > Department of Sociology
> > > McMaster University
> > > Hamilton, Ontario
> > > Canada L8S 4M4
> > > 905-525-9140x23604
> > > http://socserv.mcmaster.ca/jfox
> > >
> > >
> > > > -Original Message-
> > > > From: [EMAIL PROTECTED]
> > > > [mailto:[EMAIL PROTECTED] On Behalf Of
> Dren Scott
> > > > Sent: Monday, April 18, 2005 12:41 PM
> > > > To: 'R-Help'
> > > > Subject: RE: [R] Pearson corelation and p-value for matrix
> > > >
> > > > Hi all,
> > > >
> > > > Thanks Andy, Mark and John for all the help. I really
> > appreciate it.
> > > > I'm new to both R and statistics, so please excuse any
> > gaffes on my
> > > > part.
> > > >
> > > > Essentially what I'm trying to do, is to evaluate for
> > each row, how
> > > > many other rows would have a p-value < 0.05. So, after I
> > get my N x
> > > > N p-value matrix, I'll just filter out values that are > 0.05.
> > > >
> > > > Each of my datasets (6000 rows x 100 columns) would
> > consist of some
> > > > NA's. The iterative procedure (cor.pvalues) proposed by
> > John would
> > > > yield the values, but it would take an inordinately
> long time (I
> > > > have 50 of these datasets to process). The solution proposed by
> > > > Andy, although fast, would not be able to incorporate the NA's.
> > > >
> > > > Is there any workaround for the NA's? Or possibly do
> you think I
> > > > could try something else?
> > > >
> > > > Thanks very much.
> > > >
> > > > Dren
> > > >
> > > >
> > > > "Liaw, Andy" <[EMAIL PROTECTED]> wrote:
> > > > > From: John Fox
> > > > >
> > > > > Dear Andy,
> > > > >
> > > > > That's clearly much better -- and illustrates an
> > > effective strategy
> > > > > for vectorizing (or "matricizing") a computation. I think
> > > I'll
[R] standard normal gaussian
Hello List, I need to calculate the p-value (in a standard normal gaussian) for a Z value I have. I guess if I had to calculate by myself the value of the area in my function or refer to some already made function in R. Any hints? Thank you very much, Simone __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Re-release of R 2.1.0
Due to a mishap in the build procedure, the originally released sources erroneously had "2.2.0" in several places within the "configure" file. A re-released version has been put up in http://biostat.ku.dk/~pd/R-release/ and should find its way to CRAN and its mirrors some time tomorrow. The new files have md5 sums as follows: 94d55d512a9ba36caa9b7df079bae19f COPYING d8045f3b8f929c1cb29a1e3fd737b499 COPYING.LIB 70447ae7f2c35233d3065b004aa4f331 INSTALL 8b78e12b100a6834fdada2dfaab98ab0 NEWS 88bbd6781faedc788a1cbd434194480c ONEWS 4f004de59e24a52d0f500063b4603bcb OONEWS fb47b1fdef4323031e24d541a2f36b2b R-2.0.1.tar.gz 270f7a7382e8cb10a353148598f91096 R-2.1.0.tar.gz d8a2d5461c16aa37b1418c0500423bec R-2.1.0.tar.gz-split.aa 9a8e7b5988fc4e43b56561cbb6853534 R-2.1.0.tar.gz-split.ab dfb0600d8726613ac5a424bbd054c243 R-2.1.0.tar.gz-split.ac 150903071850a951722713560834029a R-2.1.0.tar.gz-split.ad cb416f351d0c9c75a86c27998d691372 R-2.1.0.tar.gz-split.ae d8e01f2b68d2f689444901ae1957c396 R-2.1.0.tar.gz-split.af 3949c2649d595683fe26212dc292f4b1 R-2.1.0.tar.gz-split.ag 4e984f0b2dacf5d52657b04e874756d9 R-2.1.0.tar.gz-split.ah c0f6e6135e54bd5d8917c804ec4b9b15 R-2.1.0.tar.gz-split.ai 270f7a7382e8cb10a353148598f91096 R-latest.tar.gz 56a780cdec835c5c598f8dfc0738f7f3 README AFAIK, the Windows build doesn't use "configure", so should be OK. Source builds get (at least) Makeconf wrong, causing the following symptom: > library(MASS) Warning message: package 'MASS' was built under R version 2.2.0 In that case, you'll need to rebuild with the corrected "configure". You can get the file separately from https://svn.r-project.org/R/tags/R-2-1-0/configure Apologies for the inconvenience. -pd -- O__ Peter Dalgaard Blegdamsvej 3 c/ /'_ --- Dept. of Biostatistics 2200 Cph. N (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 ___ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-announce __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Re: longer object length, is not a multiple of shorter object length in: kappa * gcounts
Hui Han wrote: Hi, Another phenomenon is that if I don't sort the data vector x3, I get the same error regardless of the bandwidth value: Actually for really large bandwidth,e.g. 120, it deosn't report errors. Sorry for the confusing description. "longer object length, is not a multiple of shorter object length in: kappa * gcounts ". Thanks in advance for any help you can give me! Hui Hui Han wrote: Hi, I was using a density estimation function as follows: > est <- KernSmooth::bkde(x3, bandwidth=10) When setting bandwidth less than 5, I got the error "longer object length, is not a multiple of shorter object length in: kappa * gcounts ". I wonder if there is anybody who can explain the error for me? Thanks! Hui __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Re: longer object length, is not a multiple of shorter object length in: kappa * gcounts
Hi, Another phenomenon is that if I don't sort the data vector x3, I get the same error regardless of the bandwidth value: "longer object length, is not a multiple of shorter object length in: kappa * gcounts ". Thanks in advance for any help you can give me! Hui Hui Han wrote: Hi, I was using a density estimation function as follows: > est <- KernSmooth::bkde(x3, bandwidth=10) When setting bandwidth less than 5, I got the error "longer object length, is not a multiple of shorter object length in: kappa * gcounts ". I wonder if there is anybody who can explain the error for me? Thanks! Hui __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] longer object length, is not a multiple of shorter object length in: kappa * gcounts
Hi, I was using a density estimation function as follows: > est <- KernSmooth::bkde(x3, bandwidth=10) When setting bandwidth less than 5, I got the error "longer object length, is not a multiple of shorter object length in: kappa * gcounts ". I wonder if there is anybody who can explain the error for me? Thanks! Hui __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Pearson corelation and p-value for matrix
Dear Andy,
Very nice! (My point was that if this is a one-time thing, for Dren to
puzzle over it is probably more time-consuming than simply doing it
inefficiently.)
Regards,
John
John Fox
Department of Sociology
McMaster University
Hamilton, Ontario
Canada L8S 4M4
905-525-9140x23604
http://socserv.mcmaster.ca/jfox
> -Original Message-
> From: Liaw, Andy [mailto:[EMAIL PROTECTED]
> Sent: Monday, April 18, 2005 2:40 PM
> To: 'John Fox'; 'Dren Scott'
> Cc: 'R-Help'
> Subject: RE: [R] Pearson corelation and p-value for matrix
>
> I believe this will do:
>
> cor.pval2 <- function(x, alternative="two-sided") {
> corMat <- cor(x, use=if (any(is.na(x)))
> "pairwise.complete.obs" else
> "all")
> df <- crossprod(!is.na(x)) - 2
> STATISTIC <- sqrt(df) * corMat / sqrt(1 - corMat^2)
> p <- pt(STATISTIC, df)
> p <- if (alternative == "less") {
> p
> } else if (alternative == "greater") {
> 1 - p
> } else 2 * pmin(p, 1 - p)
> p
> }
>
> Some test:
>
> > set.seed(1)
> > x <- matrix(runif(2e3 * 1e2), 2e3)
> > system.time(res1 <- cor.pval(t(x)), gcFirst=TRUE)
> [1] 17.28 0.77 18.16NANA
> > system.time(res2 <- cor.pval2(t(x)), gcFirst=TRUE)
> [1] 19.51 1.05 20.70NANA
> > max(abs(res1 - res2))
> [1] 0
> > x[c(1, 3, 6), c(2, 4)] <- NA
> > x[30, 61] <- NA
> > system.time(res2 <- cor.pval2(t(x)), gcFirst=TRUE)
> [1] 24.48 0.71 25.28NANA
>
> This is a bit slower because of the extra computation for
> "df". One can try to save some computation by only computing
> with the lower (or upper) triangular part.
>
> Cheers,
> Andy
>
> > From: John Fox
> >
> > Dear Dren,
> >
> > Since cor(), on which Andy's solution is based, can compute
> > pairwise-present correlations, you could adapt his function
> -- you'll
> > have to adjust the df for each pair. Alternatively, you
> could probably
> > save some time (programming time + computer time) by just using my
> > solution:
> >
> > > R <- diag(100)
> > > R[upper.tri(R)] <- R[lower.tri(R)] <- .5
> > > library(mvtnorm)
> > > X <- rmvnorm(6000, sigma=R)
> > > system.time(for (i in 1:50) cor.pvalues(X), gc=TRUE)
> > [1] 518.19 1.11 520.23 NA NA
> >
> > I know that time is money, but nine minutes (on my machine)
> probably
> > won't bankrupt anyone.
> >
> > Regards,
> > John
> >
> >
> > John Fox
> > Department of Sociology
> > McMaster University
> > Hamilton, Ontario
> > Canada L8S 4M4
> > 905-525-9140x23604
> > http://socserv.mcmaster.ca/jfox
> >
> >
> > > -Original Message-
> > > From: [EMAIL PROTECTED]
> > > [mailto:[EMAIL PROTECTED] On Behalf Of Dren Scott
> > > Sent: Monday, April 18, 2005 12:41 PM
> > > To: 'R-Help'
> > > Subject: RE: [R] Pearson corelation and p-value for matrix
> > >
> > > Hi all,
> > >
> > > Thanks Andy, Mark and John for all the help. I really
> appreciate it.
> > > I'm new to both R and statistics, so please excuse any
> gaffes on my
> > > part.
> > >
> > > Essentially what I'm trying to do, is to evaluate for
> each row, how
> > > many other rows would have a p-value < 0.05. So, after I
> get my N x
> > > N p-value matrix, I'll just filter out values that are > 0.05.
> > >
> > > Each of my datasets (6000 rows x 100 columns) would
> consist of some
> > > NA's. The iterative procedure (cor.pvalues) proposed by
> John would
> > > yield the values, but it would take an inordinately long time (I
> > > have 50 of these datasets to process). The solution proposed by
> > > Andy, although fast, would not be able to incorporate the NA's.
> > >
> > > Is there any workaround for the NA's? Or possibly do you think I
> > > could try something else?
> > >
> > > Thanks very much.
> > >
> > > Dren
> > >
> > >
> > > "Liaw, Andy" <[EMAIL PROTECTED]> wrote:
> > > > From: John Fox
> > > >
> > > > Dear Andy,
> > > >
> > > > That's clearly much better -- and illustrates an
> > effective strategy
> > > > for vectorizing (or "matricizing") a computation. I think
> > I'll add
> > > > this to my list of programming examples. It might be a little
> > > > dangerous to pass ...
> > > > through to cor(), since someone could specify
> > type="spearman", for
> > > > example.
> > >
> > > Ah, yes, the "..." isn't likely to help here! Also, it will only
> > > work correctly if there are no NA's, for example (or else
> the degree
> > > of freedom would be wrong).
> > >
> > > Best,
> > > Andy
> > >
> > > > Thanks,
> > > > John
> > > >
> > > >
> > > > John Fox
> > > > Department of Sociology
> > > > McMaster University
> > > > Hamilton, Ontario
> > > > Canada L8S 4M4
> > > > 905-525-9140x23604
> > > > http://socserv.mcmaster.ca/jfox
> > > >
> > > >
> > > > > -Original Message-
> > > > > From: Liaw, Andy [mailto:[EMAIL PROTECTE
[R] Discrepancy between gam from gam package and gam in S-PLUS
Dear Trevor, I've noticed a discrepancy in the degrees of freedom reported by gam() from the gam package in R vs. gam() in S-PLUS. The nonparametric df differ by 1; otherwise (except for things that depend upon the df), the output is the same: - snip *** From R (gam version 0.93): > mod.gam <- gam(prestige ~ lo(income, span=.6), data=Prestige) > summary(mod.gam) Call: gam(formula = prestige ~ lo(income, span = 0.6), data = Prestige) Deviance Residuals: Min 1Q Median 3Q Max -17.163 -8.205 -1.998 8.070 32.326 (Dispersion Parameter for gaussian family taken to be 122.5047) Null Deviance: 29895.43 on 101 degrees of freedom Residual Deviance: 12004.72 on 97.9939 degrees of freedom AIC: 785.82 Number of Local Scoring Iterations: 2 DF for Terms and F-values for Nonparametric Effects Df Npar Df Npar F Pr(F) (Intercept) 1 lo(income, span = 0.6) 1 2 10.626 6.537e-05 *** --- Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1 *** From S-PLUS 6.2.1 for Windows: > library(car) > mod.gam <- gam(prestige ~ lo(income, span=.6), data=Prestige) > summary(mod.gam) Call: gam(formula = prestige ~ lo(income, span = 0.6), data = Prestige) Deviance Residuals: Min1QMedian 3Q Max -17.16264 -8.205246 -1.997525 8.070375 32.32608 (Dispersion Parameter for Gaussian family taken to be 123.7677 ) Null Deviance: 29895.43 on 101 degrees of freedom Residual Deviance: 12004.72 on 96.99392 degrees of freedom Number of Local Scoring Iterations: 1 DF for Terms and F-values for Nonparametric Effects Df Npar Df Npar FPr(F) (Intercept) 1 lo(income, span = 0.6) 1 3 7.018995 0.0002517358 > - snip I suppose that one of these must be in error. Regards, John John Fox Department of Sociology McMaster University Hamilton, Ontario Canada L8S 4M4 905-525-9140x23604 http://socserv.mcmaster.ca/jfox __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] R-squared in summary(lm...)
Please consult any basic book on linear regression/linear models; for example, APPLIED REGRESSION ANALYSIS by Draper and Smith. Or google on "adjusted R-Squared." In brief, Adjusted R-squared arrempts to compensate for the property that adding extra regressors -- even random ones -- to a linear model **always** increases R-squared. It does so by penalizing for extra regressors. So R-squared is almost never useful as an indication of the quality of fit. Adjusted R-squared might be (but often isn't either because of selection bias among other reasons). -- Bert Gunter Genentech Non-Clinical Statistics South San Francisco, CA "The business of the statistician is to catalyze the scientific learning process." - George E. P. Box > -Original Message- > From: [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED] On Behalf Of > Benjamin M. Osborne > Sent: Monday, April 18, 2005 3:52 PM > To: R help > Subject: [R] R-squared in summary(lm...) > > What is the difference between the two R-squareds returned > for a linear > regression by summary(lm...)? When might one report multiple > vs. adjusted > R-squared? > Thank you, > Ben Osborne > > -- > Botany Department > University of Vermont > 109 Carrigan Drive > Burlington, VT 05405 > > [EMAIL PROTECTED] > phone: 802-656-0297 > fax: 802-656-0440 > > __ > R-help@stat.math.ethz.ch mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! > http://www.R-project.org/posting-guide.html > __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] using imported tables
I've recently found out using read.table i can insert a table into R from a .dat file. The problem i'm having is now that i've got R to read the package i want to use the data from each column. I've tried adding titles to each column using col.name, but it assigns a name to the entire table and wont recognize the name i gave the table later in programming. Is there something i'm doing wrong or am i missing an important step? I'd appreciate any help. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] R-squared in summary(lm...)
What is the difference between the two R-squareds returned for a linear regression by summary(lm...)? When might one report multiple vs. adjusted R-squared? Thank you, Ben Osborne -- Botany Department University of Vermont 109 Carrigan Drive Burlington, VT 05405 [EMAIL PROTECTED] phone: 802-656-0297 fax: 802-656-0440 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Construction of a large sparse matrix
The dense blocks are too big as Reid has already written --
for smaller instances of this sort of thing I would suggest that the
the kronecker
product %x% operator in SparseM, would be more convenient.
url:www.econ.uiuc.edu/~rogerRoger Koenker
email [EMAIL PROTECTED] Department of Economics
vox:217-333-4558University of Illinois
fax:217-244-6678Champaign, IL 61820
On Apr 18, 2005, at 3:54 PM, Doran, Harold wrote:
Dear List:
I'm working to construct a very large sparse matrix and have found
relief using the SparseM package. I have encountered an issue that is
confusing to me and wonder if anyone may be able to suggest a smarter
solution. The matrix I'm creating is a covariance matrix for a larger
research problem that is subsequently used in a simulation. Below is
the
latex form of the matrix if anyone wants to see the pattern I am trying
to create.
The core of my problem seems to localize to the last line of the
following portion of code.
n<-sample.size*4
k<-n/4
vl.mat <- as.matrix.csr(0, n, n)
block <- 1:k #each submatrix size
for(i in 1:3) vl.mat[i *k + block, i*k + block] <- LE
When the variable LE is 0, the matrix is easily created. For example,
when sample.size = 10,000 this matrix was created on my machine in
about
1 second. Here is the object size.
object.size(vl.mat)
[1] 160692
However, when LE is any number other than 0, the code generates an
error. For example, when I try LE <- 2 I get
Error: cannot allocate vector of size 781250 Kb
In addition: Warning message:
Reached total allocation of 1024Mb: see help(memory.size)
Error in as.matrix.coo(as.matrix.csr(value, nrow = length(rw), ncol =
length(cl))) :
Unable to find the argument "x" in selecting a method for
function "as.matrix.coo"
I'm guessing that single digit integers should occupy the same amount
of
memory. So, I'm thinking that the matrix is "less sparse" and the
problem is related to the introduction of a non-zero element (seems
obvious). However, the matrix still retains a very large proportion of
zeros. In fact, there are still more zeros than non-zero elements.
Can anyone suggest a reason why I am not able to create this matrix?
I'm
at the limit of my experience and could use a pointer if anyone is able
to provide one.
Many thanks,
Harold
P.S. The matrix above is added to another matrix to create the
covariance matrix below. The code above is designed to create the
portion of the matrix \sigma^2_{vle}\bm{J} .
\begin{equation}
\label{vert:cov}
\bm{\Phi} = var
\left [
\begin{array}{c}
Y^*_{1}\\
Y^*_{2}\\
Y^*_{3}\\
Y^*_{4}\\
\end{array}
\right ]
=
\left [
\begin{array}{}
\sigma^2_{\epsilon}\bm{I}& \sigma^2_{\epsilon}\rho\bm{I} & \bm{0} &
\bm{0}\\
\sigma^2_{\epsilon}\rho\bm{I} &
\sigma^2_{\epsilon}\bm{I}+\sigma^2_{vle}\bm{J} &
\sigma^2_{\epsilon}\rho^2\bm{I} & \bm{0}\\
\bm{0} & \sigma^2_{\epsilon}\rho^2\bm{I} &
\sigma^2_{\epsilon}\bm{I}+\sigma^2_{vle}\bm{J}&
\sigma^2_{\epsilon}\rho^3\bm{I}\\
\bm{0} & \bm{0} & \sigma^2_{\epsilon}\rho^3\bm{I}&
\sigma^2_{\epsilon}\bm{I}+\sigma^2_{vle}\bm{J} \\
\end{array}
\right]
\end{equation}
where $\bm{I}$ is the identity matrix, $\bm{J}$ is the unity matrix,
and
$\rho$ is the autocorrelation.
[[alternative HTML version deleted]]
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Construction of a large sparse matrix
Your statements for sample.size=10,000 try to construct a matrix with 3
dense 10,000 x 10,000 blocks. That's approximately 10*10*8 MB each and very
likely explains your error message.
Since you have a simple formula for the matrix, why not define a function to
implement multiplication by this matrix (and whatever else you want to do
with it), rather than use general sparse matrix representations?
Reid Huntsinger
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Doran, Harold
Sent: Monday, April 18, 2005 4:55 PM
To: r-help@stat.math.ethz.ch
Subject: [R] Construction of a large sparse matrix
Dear List:
I'm working to construct a very large sparse matrix and have found
relief using the SparseM package. I have encountered an issue that is
confusing to me and wonder if anyone may be able to suggest a smarter
solution. The matrix I'm creating is a covariance matrix for a larger
research problem that is subsequently used in a simulation. Below is the
latex form of the matrix if anyone wants to see the pattern I am trying
to create.
The core of my problem seems to localize to the last line of the
following portion of code.
n<-sample.size*4
k<-n/4
vl.mat <- as.matrix.csr(0, n, n)
block <- 1:k #each submatrix size
for(i in 1:3) vl.mat[i *k + block, i*k + block] <- LE
When the variable LE is 0, the matrix is easily created. For example,
when sample.size = 10,000 this matrix was created on my machine in about
1 second. Here is the object size.
> object.size(vl.mat)
[1] 160692
However, when LE is any number other than 0, the code generates an
error. For example, when I try LE <- 2 I get
Error: cannot allocate vector of size 781250 Kb
In addition: Warning message:
Reached total allocation of 1024Mb: see help(memory.size)
Error in as.matrix.coo(as.matrix.csr(value, nrow = length(rw), ncol =
length(cl))) :
Unable to find the argument "x" in selecting a method for
function "as.matrix.coo"
I'm guessing that single digit integers should occupy the same amount of
memory. So, I'm thinking that the matrix is "less sparse" and the
problem is related to the introduction of a non-zero element (seems
obvious). However, the matrix still retains a very large proportion of
zeros. In fact, there are still more zeros than non-zero elements.
Can anyone suggest a reason why I am not able to create this matrix? I'm
at the limit of my experience and could use a pointer if anyone is able
to provide one.
Many thanks,
Harold
P.S. The matrix above is added to another matrix to create the
covariance matrix below. The code above is designed to create the
portion of the matrix \sigma^2_{vle}\bm{J} .
\begin{equation}
\label{vert:cov}
\bm{\Phi} = var
\left [
\begin{array}{c}
Y^*_{1}\\
Y^*_{2}\\
Y^*_{3}\\
Y^*_{4}\\
\end{array}
\right ]
=
\left [
\begin{array}{}
\sigma^2_{\epsilon}\bm{I}& \sigma^2_{\epsilon}\rho\bm{I} & \bm{0} &
\bm{0}\\
\sigma^2_{\epsilon}\rho\bm{I} &
\sigma^2_{\epsilon}\bm{I}+\sigma^2_{vle}\bm{J} &
\sigma^2_{\epsilon}\rho^2\bm{I} & \bm{0}\\
\bm{0} & \sigma^2_{\epsilon}\rho^2\bm{I} &
\sigma^2_{\epsilon}\bm{I}+\sigma^2_{vle}\bm{J}&
\sigma^2_{\epsilon}\rho^3\bm{I}\\
\bm{0} & \bm{0} & \sigma^2_{\epsilon}\rho^3\bm{I}&
\sigma^2_{\epsilon}\bm{I}+\sigma^2_{vle}\bm{J} \\
\end{array}
\right]
\end{equation}
where $\bm{I}$ is the identity matrix, $\bm{J}$ is the unity matrix, and
$\rho$ is the autocorrelation.
[[alternative HTML version deleted]]
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] R 2.1.0 GUI language
On Mon, 18 Apr 2005, Stephan Tolksdorf wrote: [...] LC_ALL doesn't seem to change the message language. LC_MESSAGES, which is mentioned in the installation and administration guide, doesn't seem to exist in R 2.1.0 and isn't listed in the locales help. Right: it cannot be set inside R. For as the manual says: Note that you should not expect to be able to change the language once R is running. But it can be set as an environment variable outside R, as described in the rw-FAQ. Setting either LC_MESSAGES or LC_ALL=de on the command line works for me. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Construction of a large sparse matrix
Dear List:
I'm working to construct a very large sparse matrix and have found
relief using the SparseM package. I have encountered an issue that is
confusing to me and wonder if anyone may be able to suggest a smarter
solution. The matrix I'm creating is a covariance matrix for a larger
research problem that is subsequently used in a simulation. Below is the
latex form of the matrix if anyone wants to see the pattern I am trying
to create.
The core of my problem seems to localize to the last line of the
following portion of code.
n<-sample.size*4
k<-n/4
vl.mat <- as.matrix.csr(0, n, n)
block <- 1:k #each submatrix size
for(i in 1:3) vl.mat[i *k + block, i*k + block] <- LE
When the variable LE is 0, the matrix is easily created. For example,
when sample.size = 10,000 this matrix was created on my machine in about
1 second. Here is the object size.
> object.size(vl.mat)
[1] 160692
However, when LE is any number other than 0, the code generates an
error. For example, when I try LE <- 2 I get
Error: cannot allocate vector of size 781250 Kb
In addition: Warning message:
Reached total allocation of 1024Mb: see help(memory.size)
Error in as.matrix.coo(as.matrix.csr(value, nrow = length(rw), ncol =
length(cl))) :
Unable to find the argument "x" in selecting a method for
function "as.matrix.coo"
I'm guessing that single digit integers should occupy the same amount of
memory. So, I'm thinking that the matrix is "less sparse" and the
problem is related to the introduction of a non-zero element (seems
obvious). However, the matrix still retains a very large proportion of
zeros. In fact, there are still more zeros than non-zero elements.
Can anyone suggest a reason why I am not able to create this matrix? I'm
at the limit of my experience and could use a pointer if anyone is able
to provide one.
Many thanks,
Harold
P.S. The matrix above is added to another matrix to create the
covariance matrix below. The code above is designed to create the
portion of the matrix \sigma^2_{vle}\bm{J} .
\begin{equation}
\label{vert:cov}
\bm{\Phi} = var
\left [
\begin{array}{c}
Y^*_{1}\\
Y^*_{2}\\
Y^*_{3}\\
Y^*_{4}\\
\end{array}
\right ]
=
\left [
\begin{array}{}
\sigma^2_{\epsilon}\bm{I}& \sigma^2_{\epsilon}\rho\bm{I} & \bm{0} &
\bm{0}\\
\sigma^2_{\epsilon}\rho\bm{I} &
\sigma^2_{\epsilon}\bm{I}+\sigma^2_{vle}\bm{J} &
\sigma^2_{\epsilon}\rho^2\bm{I} & \bm{0}\\
\bm{0} & \sigma^2_{\epsilon}\rho^2\bm{I} &
\sigma^2_{\epsilon}\bm{I}+\sigma^2_{vle}\bm{J}&
\sigma^2_{\epsilon}\rho^3\bm{I}\\
\bm{0} & \bm{0} & \sigma^2_{\epsilon}\rho^3\bm{I}&
\sigma^2_{\epsilon}\bm{I}+\sigma^2_{vle}\bm{J} \\
\end{array}
\right]
\end{equation}
where $\bm{I}$ is the identity matrix, $\bm{J}$ is the unity matrix, and
$\rho$ is the autocorrelation.
[[alternative HTML version deleted]]
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Very Slow Gower Similarity Function
Quoting Martin Maechler <[EMAIL PROTECTED]>:
> I don't know what exactly you want.
The Gower coefficient I am referring to comes from his 1971 article in
Biometrics (27(4):857-871). It differs from most commonly used measures (but
not, apparently, daisy!) by allowing the incorporation of quantitative and
qualitative (binary or unordered multistate characters) variables, and also by
providing a mechanism for dropping missing values from similarity calculations.
This is also covered in Legendre and Legendre.
>
> The function daisy() in the recommended package "cluster"
> has always worked with missing values and IIRC, the book
> "Kaufman & Rousseeuw" {which I have not at hand here at home},
> clearly mentions Gower's origin of their distance measure
> definition.
I was unaware of the daisy function. Looking over it now it differs from the
Gower coefficient primarily in the method of standardization. Gower
standardized each variable by dividing it by it's range ("ranging"), where
daisy does a more conventional standardization (-mean and /SD). As I understand
it, there isn't much to recommend standardizing over ranging (or vice versa) so
daisy may provide a useful alternative for my project. I'll have to look into
it!
Thanks,
Tyler
>
> Martin Maechler, maintainer of cluster package,
> ETH Zurich
>
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] R 2.1.0 GUI language
Thanks for your prompt reply.
Which OS is this? PLEASE do read the posting guide.
Win XP, SP2 (German), which I forgot to mention despite heaving read the
guide.
This *is* documented in the `R Installation and Adminstration' Manual,
section `Internationalization'.
Looked in the wrong place (and the help isn't globally searchable).
If your OS is set to German, then why don't you expect R to follow?
I'd just like to have an option to change the language. English is my
working language and the common denominator of my tool chain. German
warning messages probably also wouldn't be much help to me when asking
for help on this list.
By the way, I'd be quite afraid to suddenly see the German decimal comma
instead of point in my program output, but fortunately R isn't that
localized.
If you don't want German menus, do you want German months (as has
happened for many versions).
For Windows or Linux, to just change the language of messages, set
environment variable LANGUAGE=en. To set all aspects, set LC_ALL=en. I
don't know about MacOS X menus.
LC_ALL doesn't seem to change the message language. LC_MESSAGES,
which is mentioned in the installation and administration guide,
doesn't seem to exist in R 2.1.0 and isn't listed in the locales help.
I helped myself by putting
Sys.putenv("LANGUAGE"="EN");Sys.setlocale("LC_ALL","EN")
into etc/Rprofile. Don't know if this is the default way to go.
Stephan
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Pearson corelation and p-value for matrix
I believe this will do:
cor.pval2 <- function(x, alternative="two-sided") {
corMat <- cor(x, use=if (any(is.na(x))) "pairwise.complete.obs" else
"all")
df <- crossprod(!is.na(x)) - 2
STATISTIC <- sqrt(df) * corMat / sqrt(1 - corMat^2)
p <- pt(STATISTIC, df)
p <- if (alternative == "less") {
p
} else if (alternative == "greater") {
1 - p
} else 2 * pmin(p, 1 - p)
p
}
Some test:
> set.seed(1)
> x <- matrix(runif(2e3 * 1e2), 2e3)
> system.time(res1 <- cor.pval(t(x)), gcFirst=TRUE)
[1] 17.28 0.77 18.16NANA
> system.time(res2 <- cor.pval2(t(x)), gcFirst=TRUE)
[1] 19.51 1.05 20.70NANA
> max(abs(res1 - res2))
[1] 0
> x[c(1, 3, 6), c(2, 4)] <- NA
> x[30, 61] <- NA
> system.time(res2 <- cor.pval2(t(x)), gcFirst=TRUE)
[1] 24.48 0.71 25.28NANA
This is a bit slower because of the extra computation for "df". One can try
to save some computation by only computing with the lower (or upper)
triangular part.
Cheers,
Andy
> From: John Fox
>
> Dear Dren,
>
> Since cor(), on which Andy's solution is based, can compute
> pairwise-present
> correlations, you could adapt his function -- you'll have to
> adjust the df
> for each pair. Alternatively, you could probably save some
> time (programming
> time + computer time) by just using my solution:
>
> > R <- diag(100)
> > R[upper.tri(R)] <- R[lower.tri(R)] <- .5
> > library(mvtnorm)
> > X <- rmvnorm(6000, sigma=R)
> > system.time(for (i in 1:50) cor.pvalues(X), gc=TRUE)
> [1] 518.19 1.11 520.23 NA NA
>
> I know that time is money, but nine minutes (on my machine)
> probably won't
> bankrupt anyone.
>
> Regards,
> John
>
>
> John Fox
> Department of Sociology
> McMaster University
> Hamilton, Ontario
> Canada L8S 4M4
> 905-525-9140x23604
> http://socserv.mcmaster.ca/jfox
>
>
> > -Original Message-
> > From: [EMAIL PROTECTED]
> > [mailto:[EMAIL PROTECTED] On Behalf Of Dren Scott
> > Sent: Monday, April 18, 2005 12:41 PM
> > To: 'R-Help'
> > Subject: RE: [R] Pearson corelation and p-value for matrix
> >
> > Hi all,
> >
> > Thanks Andy, Mark and John for all the help. I really
> > appreciate it. I'm new to both R and statistics, so please
> > excuse any gaffes on my part.
> >
> > Essentially what I'm trying to do, is to evaluate for each
> > row, how many other rows would have a p-value < 0.05. So,
> > after I get my N x N p-value matrix, I'll just filter out
> > values that are > 0.05.
> >
> > Each of my datasets (6000 rows x 100 columns) would consist
> > of some NA's. The iterative procedure (cor.pvalues) proposed
> > by John would yield the values, but it would take an
> > inordinately long time (I have 50 of these datasets to
> > process). The solution proposed by Andy, although fast, would
> > not be able to incorporate the NA's.
> >
> > Is there any workaround for the NA's? Or possibly do you
> > think I could try something else?
> >
> > Thanks very much.
> >
> > Dren
> >
> >
> > "Liaw, Andy" <[EMAIL PROTECTED]> wrote:
> > > From: John Fox
> > >
> > > Dear Andy,
> > >
> > > That's clearly much better -- and illustrates an
> effective strategy
> > > for vectorizing (or "matricizing") a computation. I think
> I'll add
> > > this to my list of programming examples. It might be a little
> > > dangerous to pass ...
> > > through to cor(), since someone could specify
> type="spearman", for
> > > example.
> >
> > Ah, yes, the "..." isn't likely to help here! Also, it will
> > only work correctly if there are no NA's, for example (or
> > else the degree of freedom would be wrong).
> >
> > Best,
> > Andy
> >
> > > Thanks,
> > > John
> > >
> > >
> > > John Fox
> > > Department of Sociology
> > > McMaster University
> > > Hamilton, Ontario
> > > Canada L8S 4M4
> > > 905-525-9140x23604
> > > http://socserv.mcmaster.ca/jfox
> > >
> > >
> > > > -Original Message-
> > > > From: Liaw, Andy [mailto:[EMAIL PROTECTED]
> > > > Sent: Friday, April 15, 2005 9:51 PM
> > > > To: 'John Fox'; [EMAIL PROTECTED]
> > > > Cc: 'R-Help'; 'Dren Scott'
> > > > Subject: RE: [R] Pearson corelation and p-value for matrix
> > > >
> > > > We can be a bit sneaky and `borrow' code from cor.test.default:
> > > >
> > > > cor.pval <- function(x, alternative="two-sided", ...) {
> corMat <-
> > > > cor(x, ...) n <- nrow(x) df <- n - 2 STATISTIC <-
> > sqrt(df) * corMat
> > > > / sqrt(1 - corMat^2) p <- pt(STATISTIC, df) p <- if
> > (alternative ==
> > > > "less") { p } else if (alternative == "greater") {
> > > > 1 - p
> > > > } else 2 * pmin(p, 1 - p)
> > > > p
> > > > }
> > > >
> > > > The test:
> > > >
> > > > > system.time(c1 <- cor.pvals(X), gcFirst=TRUE)
> > > > [1] 13.19 0.01 13.58 NA NA
> > > > > system.time(c2 <- cor.pvalues(X), gcFirst=TRUE)
> > > > [1] 6.22 0.00 6.42 NA NA
> > > > > system.time(c3 <- cor.p
RE: [R] when can we expect Prof Tierney's compiled R?
Do you mean the byte code compiler? You can find it at: http://www.stat.uiowa.edu/~luke/R/compiler/ Andy > From: Jason Liao > > I am excited to learn that Prof. Tierney is bringing to us compiled R. > I would like to learn when it will be available. This information will > be useful in scheduling some of my projects. Thanks. > > Jason > > > Jason Liao, http://www.geocities.com/jg_liao > Dept. of Biostatistics, http://www2.umdnj.edu/bmtrxweb > University of Medicine and Dentistry of New Jersey > 683 Hoes Lane West, Piscataway' NJ 08854 > phone 732-235-5429, School of Public Health office > phone 732-235-9824, Cancer Institute of New Jersey office > > __ > R-help@stat.math.ethz.ch mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! > http://www.R-project.org/posting-guide.html > > > __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Very Slow Gower Similarity Function
> "Tyler" == Tyler Smith <[EMAIL PROTECTED]>
> on Mon, 18 Apr 2005 12:10:34 -0400 writes:
Tyler> Hello, I am a relatively new user of R. I have
Tyler> written a basic function to calculate the Gower
Tyler> similarity function. I was motivated to do so partly
Tyler> as an excercise in learning R, and partly because the
Tyler> existing option (vegdist in the vegan package) does
Tyler> not accept missing values.
I don't know what exactly you want.
The function daisy() in the recommended package "cluster"
has always worked with missing values and IIRC, the book
"Kaufman & Rousseeuw" {which I have not at hand here at home},
clearly mentions Gower's origin of their distance measure
definition.
Martin Maechler, maintainer of cluster package,
ETH Zurich
Tyler> I think I have succeeded - my function gives me the
Tyler> correct values. However, now that I'm starting to use
Tyler> it with real data, I realise it's very slow. It takes
Tyler> more than 45 minutes on my Windows 98 machine (R
Tyler> 2.0.1 Patched (2005-03-29)) with a 185x32 matrix with
Tyler> ca 100 missing values. If anyone can suggest ways to
Tyler> speed up my function I would appreciate it. I suspect
Tyler> having a pair of nested for loops is the problem, but
Tyler> I couldn't figure out how to get rid of them.
Tyler> The function is:
Tyler> ### Gower Similarity Matrix###
Tyler> sGow <- function (mat){
Tyler> OBJ <- nrow(mat) #number of objects MATDESC <- ncol
Tyler> (mat) #number of descriptors MRANGE <- apply
Tyler> (mat,2,max, na.rm=T)-apply (mat,2,min,na.rm=T) #descr
Tyler> ranges DESCRIPT <- 1:MATDESC #descriptor index vector
Tyler> smat <- matrix(1, nrow = OBJ, ncol = OBJ) #'empty'
Tyler> similarity matrix
Tyler> for (i in 1:OBJ){ for (j in i:OBJ){
Tyler> ##calculate index vector of non-NA descriptors
Tyler> between objects i and j descvect <- intersect
Tyler> (setdiff (DESCRIPT,
Tyler> DESCRIPT[is.na(mat[i,DESCRIPT])]), setdiff (DESCRIPT,
Tyler> DESCRIPT[is.na (mat[j,DESCRIPT])]))
Tyler> descnum <- length(descvect) # number of valid
Tyler> descr for i~j comparison
Tyler> partialsim <- (1-
Tyler> abs(mat[i,descvect]-mat[j,descvect])/MRANGE[descvect])
Tyler> smat[i,j] <- smat[j,i] <- sum (partialsim) /
Tyler> descnum } } smat }
Tyler> Thank-you for your time,
Tyler> Tyler
Tyler> -- Tyler Smith
Tyler> PhD Candidate Plant Science Department McGill
Tyler> University
Tyler> [EMAIL PROTECTED]
Tyler> __
Tyler> R-help@stat.math.ethz.ch mailing list
Tyler> https://stat.ethz.ch/mailman/listinfo/r-help PLEASE
Tyler> do read the posting guide!
Tyler> http://www.R-project.org/posting-guide.html
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] R 2.1.0 GUI language
On Mon, 18 Apr 2005, Stephan Tolksdorf wrote: How do I set the language of the GUI? With R 2.1.0 it suddenly changed to German and I want it to stay English. I couldn't find any setting to change. Which OS is this? PLEASE do read the posting guide. This *is* documented in the `R Installation and Adminstration' Manual, section `Internationalization'. If your OS is set to German, then why don't you expect R to follow? If you don't want German menus, do you want German months (as has happened for many versions). For Windows or Linux, to just change the language of messages, set environment variable LANGUAGE=en. To set all aspects, set LC_ALL=en. I don't know about MacOS X menus. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Pearson corelation and p-value for matrix
Dear Dren,
Since cor(), on which Andy's solution is based, can compute pairwise-present
correlations, you could adapt his function -- you'll have to adjust the df
for each pair. Alternatively, you could probably save some time (programming
time + computer time) by just using my solution:
> R <- diag(100)
> R[upper.tri(R)] <- R[lower.tri(R)] <- .5
> library(mvtnorm)
> X <- rmvnorm(6000, sigma=R)
> system.time(for (i in 1:50) cor.pvalues(X), gc=TRUE)
[1] 518.19 1.11 520.23 NA NA
I know that time is money, but nine minutes (on my machine) probably won't
bankrupt anyone.
Regards,
John
John Fox
Department of Sociology
McMaster University
Hamilton, Ontario
Canada L8S 4M4
905-525-9140x23604
http://socserv.mcmaster.ca/jfox
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] On Behalf Of Dren Scott
> Sent: Monday, April 18, 2005 12:41 PM
> To: 'R-Help'
> Subject: RE: [R] Pearson corelation and p-value for matrix
>
> Hi all,
>
> Thanks Andy, Mark and John for all the help. I really
> appreciate it. I'm new to both R and statistics, so please
> excuse any gaffes on my part.
>
> Essentially what I'm trying to do, is to evaluate for each
> row, how many other rows would have a p-value < 0.05. So,
> after I get my N x N p-value matrix, I'll just filter out
> values that are > 0.05.
>
> Each of my datasets (6000 rows x 100 columns) would consist
> of some NA's. The iterative procedure (cor.pvalues) proposed
> by John would yield the values, but it would take an
> inordinately long time (I have 50 of these datasets to
> process). The solution proposed by Andy, although fast, would
> not be able to incorporate the NA's.
>
> Is there any workaround for the NA's? Or possibly do you
> think I could try something else?
>
> Thanks very much.
>
> Dren
>
>
> "Liaw, Andy" <[EMAIL PROTECTED]> wrote:
> > From: John Fox
> >
> > Dear Andy,
> >
> > That's clearly much better -- and illustrates an effective strategy
> > for vectorizing (or "matricizing") a computation. I think I'll add
> > this to my list of programming examples. It might be a little
> > dangerous to pass ...
> > through to cor(), since someone could specify type="spearman", for
> > example.
>
> Ah, yes, the "..." isn't likely to help here! Also, it will
> only work correctly if there are no NA's, for example (or
> else the degree of freedom would be wrong).
>
> Best,
> Andy
>
> > Thanks,
> > John
> >
> >
> > John Fox
> > Department of Sociology
> > McMaster University
> > Hamilton, Ontario
> > Canada L8S 4M4
> > 905-525-9140x23604
> > http://socserv.mcmaster.ca/jfox
> >
> >
> > > -Original Message-
> > > From: Liaw, Andy [mailto:[EMAIL PROTECTED]
> > > Sent: Friday, April 15, 2005 9:51 PM
> > > To: 'John Fox'; [EMAIL PROTECTED]
> > > Cc: 'R-Help'; 'Dren Scott'
> > > Subject: RE: [R] Pearson corelation and p-value for matrix
> > >
> > > We can be a bit sneaky and `borrow' code from cor.test.default:
> > >
> > > cor.pval <- function(x, alternative="two-sided", ...) { corMat <-
> > > cor(x, ...) n <- nrow(x) df <- n - 2 STATISTIC <-
> sqrt(df) * corMat
> > > / sqrt(1 - corMat^2) p <- pt(STATISTIC, df) p <- if
> (alternative ==
> > > "less") { p } else if (alternative == "greater") {
> > > 1 - p
> > > } else 2 * pmin(p, 1 - p)
> > > p
> > > }
> > >
> > > The test:
> > >
> > > > system.time(c1 <- cor.pvals(X), gcFirst=TRUE)
> > > [1] 13.19 0.01 13.58 NA NA
> > > > system.time(c2 <- cor.pvalues(X), gcFirst=TRUE)
> > > [1] 6.22 0.00 6.42 NA NA
> > > > system.time(c3 <- cor.pval(X), gcFirst=TRUE)
> > > [1] 0.07 0.00 0.07 NA NA
> > >
> > > Cheers,
> > > Andy
> > >
> > > > From: John Fox
> > > >
> > > > Dear Mark,
> > > >
> > > > I think that the reflex of trying to avoid loops in R is often
> > > > mistaken, and so I decided to try to time the two
> > approaches (on a
> > > > 3GHz Windows XP system).
> > > >
> > > > I discovered, first, that there is a bug in your
> function -- you
> > > > appear to have indexed rows instead of columns; fixing that:
> > > >
> > > > cor.pvals <- function(mat)
> > > > {
> > > > cols <- expand.grid(1:ncol(mat), 1:ncol(mat))
> matrix(apply(cols,
> > > > 1,
> > > > function(x) cor.test(mat[, x[1]], mat[, x[2]])$p.value), ncol =
> > > > ncol(mat)) }
> > > >
> > > >
> > > > My function is cor.pvalues and yours cor.pvals. This is
> > for a data
> > > > matrix with 1000 observations on 100 variables:
> > > >
> > > > > R <- diag(100)
> > > > > R[upper.tri(R)] <- R[lower.tri(R)] <- .5
> > > > > library(mvtnorm)
> > > > > X <- rmvnorm(1000, sigma=R)
> > > > > dim(X)
> > > > [1] 1000 100
> > > > >
> > > > > system.time(cor.pvalues(X))
> > > > [1] 5.53 0.00 5.53 NA NA
> > > > >
> > > > > system.time(cor.pvals(X))
> > > > [1] 12.66 0.00 12.66 NA NA
> > > > >
> > > >
> > > > I frankly didn't expect the a
Re: [R] Very Slow Gower Similarity Function
On 18 Apr 2005, at 20:36, Anon. wrote: Jari Oksanen wrote: On 18 Apr 2005, at 19:10, Tyler Smith wrote: Hello, I am a relatively new user of R. I have written a basic function to calculate the Gower similarity function. I was motivated to do so partly as an excercise in learning R, and partly because the existing option (vegdist in the vegan package) does not accept missing values. Speed is the reason to use C instead of R. It should be easy, almost trivial, to modify the vegdist.c so that it handles missing values. I guess this handling means ignoring the value pair if one of the values is missing -- which is not so gentle to the metric properties so dear to Gower. Package vegan is designed for ecological community data which generally do not have missing values (except in environmental data), but contributions are welcome. The only reason you never see ecological community data with missing values is because the ecologists remove those species/sites from their Excel sheets before they give it to you to sort out their mess. Well, ecologists have plenty of missing species in their community data, but these have zero values since they were not observed. I guess some Bob O'Hara is going to have a paper about this in JAE. This is actually one of the few things they know how to do in Excel - I'm dreading the day when a paper appears in JAE saying that you can use Excel to produce P-values. The "A" in "JAE" stands for "Animal": for real things they still have Journal of Ecology. To be slightly more serious, as an exercise the OP could consider writing a wrapper function in R that removes the missing data and then calls vegdist to calculate his Gower similarity index. The looping goes within C code, and for pairwise deletion of missing values wrapping is difficult. With complete.cases this is trivial (and then your result would be more metric as well). -- Jari Oksanen, Oulu, Finland __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] when can we expect Prof Tierney's compiled R?
I am excited to learn that Prof. Tierney is bringing to us compiled R. I would like to learn when it will be available. This information will be useful in scheduling some of my projects. Thanks. Jason Jason Liao, http://www.geocities.com/jg_liao Dept. of Biostatistics, http://www2.umdnj.edu/bmtrxweb University of Medicine and Dentistry of New Jersey 683 Hoes Lane West, Piscataway NJ 08854 phone 732-235-5429, School of Public Health office phone 732-235-9824, Cancer Institute of New Jersey office __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] R 2.1.0 GUI language
Hi How do I set the language of the GUI? With R 2.1.0 it suddenly changed to German and I want it to stay English. I couldn't find any setting to change. Regards, Stephan __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Generating a binomial random variable correlated with a
Dear: Thank you very much for your 'generous' help. Overall I have learnt the limits of the solution to my problem. I was not feeling compfortable how I was going around with the problem. Bill and Ted provided a wonderful insight into what I was accomplishing and provided alternative solutions. The method suggested by Earnst and Giovanni seems to be more objective and easy to implement. Ted's follow up on Bills remark leads in the same direction too. I don't have words to thank you all. Ashraf __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] dynamic function loading
How can I do dynamic function loading in R? I have a list of R functions in the database and want to, dynamicly, execute teh function that corresponds to my in comming data. Is there a way to do that without a big if/else? __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] colClasses = "Date" in read.delim, how to pass date-format?
On Mon, 18 Apr 2005, Christoph Lehmann wrote:
so what do you recommend: I just need to be able to sort a data.frame
according to the date entry, and e.g. compute differences between subsequent
dates. Shall I stay with dates (thanks for the hint about confusion of Date
and dates) or is there a better way for this kind of task?
I suggest you use "Date": read as character and convert using as.Date
after reading.
thanks a lot
Cheers
Christoph
You are confusing class "Date" (part of R) with class "dates" (part of
package chron). There is no as() method for class "dates", so you can't
do this. You can read the column as character (not factor) and convert
later, but it sounds like the `huge amount of memory (and time)' is in
fact taken by package chron.
On Mon, 18 Apr 2005, Christoph Lehmann wrote:
I have a huge data-set with one column being of type date.
Of course I can import the data using this column as "factor" and then
convert it later to dates, using:
sws.bezuege$FaktDat <- dates(as.character(sws.bezuege$FaktDat),
format = c(dates = "d.m.y"))
But the conversion requires a huge amount of memory (and time),
therefore I
would like to use colClasses = c("Date"). My question is:
since I have format = c(dates = "d.m.y"), how can I pass this option to
read.delim(..., colClasses = c("Date")) ?
--
Brian D. Ripley, [EMAIL PROTECTED]
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UKFax: +44 1865 272595
--
+++ GMX - Die erste Adresse für Mail, Message, More +++
1 GB Mailbox bereits in GMX FreeMail http://www.gmx.net/de/go/mail
--
Brian D. Ripley, [EMAIL PROTECTED]
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UKFax: +44 1865 272595__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] colClasses = "Date" in read.delim, how to pass date-format?
so what do you recommend: I just need to be able to sort a data.frame
according to the date entry, and e.g. compute differences between subsequent
dates. Shall I stay with dates (thanks for the hint about confusion of Date
and dates) or is there a better way for this kind of task?
thanks a lot
Cheers
Christoph
> You are confusing class "Date" (part of R) with class "dates" (part of
> package chron). There is no as() method for class "dates", so you can't
> do this. You can read the column as character (not factor) and convert
> later, but it sounds like the `huge amount of memory (and time)' is in
> fact taken by package chron.
>
> On Mon, 18 Apr 2005, Christoph Lehmann wrote:
>
> > I have a huge data-set with one column being of type date.
> > Of course I can import the data using this column as "factor" and then
> > convert it later to dates, using:
> >
> > sws.bezuege$FaktDat <- dates(as.character(sws.bezuege$FaktDat),
> > format = c(dates = "d.m.y"))
> >
> >
> > But the conversion requires a huge amount of memory (and time),
> therefore I
> > would like to use colClasses = c("Date"). My question is:
> > since I have format = c(dates = "d.m.y"), how can I pass this option to
> > read.delim(..., colClasses = c("Date")) ?
>
> --
> Brian D. Ripley, [EMAIL PROTECTED]
> Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
> University of Oxford, Tel: +44 1865 272861 (self)
> 1 South Parks Road, +44 1865 272866 (PA)
> Oxford OX1 3TG, UKFax: +44 1865 272595
>
--
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Pearson corelation and p-value for matrix
Hi all,
Thanks Andy, Mark and John for all the help. I really appreciate it. I'm new to
both R and statistics, so please excuse any gaffes on my part.
Essentially what I'm trying to do, is to evaluate for each row, how many other
rows would have a p-value < 0.05. So, after I get my N x N p-value matrix, I'll
just filter out values that are > 0.05.
Each of my datasets (6000 rows x 100 columns) would consist of some NA's. The
iterative procedure (cor.pvalues) proposed by John would yield the values, but
it would take an inordinately long time (I have 50 of these datasets to
process). The solution proposed by Andy, although fast, would not be able to
incorporate the NA's.
Is there any workaround for the NA's? Or possibly do you think I could try
something else?
Thanks very much.
Dren
"Liaw, Andy" <[EMAIL PROTECTED]> wrote:
> From: John Fox
>
> Dear Andy,
>
> That's clearly much better -- and illustrates an effective
> strategy for
> vectorizing (or "matricizing") a computation. I think I'll
> add this to my
> list of programming examples. It might be a little dangerous
> to pass ...
> through to cor(), since someone could specify
> type="spearman", for example.
Ah, yes, the "..." isn't likely to help here! Also, it will
only work correctly if there are no NA's, for example (or
else the degree of freedom would be wrong).
Best,
Andy
> Thanks,
> John
>
>
> John Fox
> Department of Sociology
> McMaster University
> Hamilton, Ontario
> Canada L8S 4M4
> 905-525-9140x23604
> http://socserv.mcmaster.ca/jfox
>
>
> > -Original Message-
> > From: Liaw, Andy [mailto:[EMAIL PROTECTED]
> > Sent: Friday, April 15, 2005 9:51 PM
> > To: 'John Fox'; [EMAIL PROTECTED]
> > Cc: 'R-Help'; 'Dren Scott'
> > Subject: RE: [R] Pearson corelation and p-value for matrix
> >
> > We can be a bit sneaky and `borrow' code from cor.test.default:
> >
> > cor.pval <- function(x, alternative="two-sided", ...) {
> > corMat <- cor(x, ...)
> > n <- nrow(x)
> > df <- n - 2
> > STATISTIC <- sqrt(df) * corMat / sqrt(1 - corMat^2)
> > p <- pt(STATISTIC, df)
> > p <- if (alternative == "less") {
> > p
> > } else if (alternative == "greater") {
> > 1 - p
> > } else 2 * pmin(p, 1 - p)
> > p
> > }
> >
> > The test:
> >
> > > system.time(c1 <- cor.pvals(X), gcFirst=TRUE)
> > [1] 13.19 0.01 13.58 NA NA
> > > system.time(c2 <- cor.pvalues(X), gcFirst=TRUE)
> > [1] 6.22 0.00 6.42 NA NA
> > > system.time(c3 <- cor.pval(X), gcFirst=TRUE)
> > [1] 0.07 0.00 0.07 NA NA
> >
> > Cheers,
> > Andy
> >
> > > From: John Fox
> > >
> > > Dear Mark,
> > >
> > > I think that the reflex of trying to avoid loops in R is often
> > > mistaken, and so I decided to try to time the two
> approaches (on a
> > > 3GHz Windows XP system).
> > >
> > > I discovered, first, that there is a bug in your function -- you
> > > appear to have indexed rows instead of columns; fixing that:
> > >
> > > cor.pvals <- function(mat)
> > > {
> > > cols <- expand.grid(1:ncol(mat), 1:ncol(mat))
> > > matrix(apply(cols, 1,
> > > function(x) cor.test(mat[, x[1]], mat[,
> > > x[2]])$p.value),
> > > ncol = ncol(mat))
> > > }
> > >
> > >
> > > My function is cor.pvalues and yours cor.pvals. This is
> for a data
> > > matrix with 1000 observations on 100 variables:
> > >
> > > > R <- diag(100)
> > > > R[upper.tri(R)] <- R[lower.tri(R)] <- .5
> > > > library(mvtnorm)
> > > > X <- rmvnorm(1000, sigma=R)
> > > > dim(X)
> > > [1] 1000 100
> > > >
> > > > system.time(cor.pvalues(X))
> > > [1] 5.53 0.00 5.53 NA NA
> > > >
> > > > system.time(cor.pvals(X))
> > > [1] 12.66 0.00 12.66 NA NA
> > > >
> > >
> > > I frankly didn't expect the advantage of my approach to be
> > this large,
> > > but there it is.
> > >
> > > Regards,
> > > John
> > >
> > >
> > > John Fox
> > > Department of Sociology
> > > McMaster University
> > > Hamilton, Ontario
> > > Canada L8S 4M4
> > > 905-525-9140x23604
> > > http://socserv.mcmaster.ca/jfox
> > >
> > >
> > > > -Original Message-
> > > > From: Marc Schwartz [mailto:[EMAIL PROTECTED]
> > > > Sent: Friday, April 15, 2005 7:08 PM
> > > > To: John Fox
> > > > Cc: 'Dren Scott'; R-Help
> > > > Subject: RE: [R] Pearson corelation and p-value for matrix
> > > >
> > > > Here is what might be a slightly more efficient way to
> > get to John's
> > > > question:
> > > >
> > > > cor.pvals <- function(mat)
> > > > {
> > > > rows <- expand.grid(1:nrow(mat), 1:nrow(mat))
> > > > matrix(apply(rows, 1,
> > > > function(x) cor.test(mat[x[1], ], mat[x[2],
> > > > ])$p.value),
> > > > ncol = nrow(mat))
> > > > }
> > > >
> > > > HTH,
> > > >
> > > > Marc Schwartz
> > > >
> > > > On Fri, 2005-04-15 at 18:26 -0400, John Fox wrote:
> > > > > Dear Dren,
> > > > >
> > > > > How about the following?
> > > > >
> > > > > cor.pvalues <- function(X){
> > > > > nc <- ncol(X)
> > > > > res <- m
Re: [R] Very Slow Gower Similarity Function
Jari Oksanen wrote: On 18 Apr 2005, at 19:10, Tyler Smith wrote: Hello, I am a relatively new user of R. I have written a basic function to calculate the Gower similarity function. I was motivated to do so partly as an excercise in learning R, and partly because the existing option (vegdist in the vegan package) does not accept missing values. Speed is the reason to use C instead of R. It should be easy, almost trivial, to modify the vegdist.c so that it handles missing values. I guess this handling means ignoring the value pair if one of the values is missing -- which is not so gentle to the metric properties so dear to Gower. Package vegan is designed for ecological community data which generally do not have missing values (except in environmental data), but contributions are welcome. The only reason you never see ecological community data with missing values is because the ecologists remove those species/sites from their Excel sheets before they give it to you to sort out their mess. This is actually one of the few things they know how to do in Excel - I'm dreading the day when a paper appears in JAE saying that you can use Excel to produce P-values. To be slightly more serious, as an exercise the OP could consider writing a wrapper function in R that removes the missing data and then calls vegdist to calculate his Gower similarity index. Bob -- Bob O'Hara Department of Mathematics and Statistics P.O. Box 68 (Gustaf Hällströmin katu 2b) FIN-00014 University of Helsinki Finland Telephone: +358-9-191 51479 Mobile: +358 50 599 0540 Fax: +358-9-191 51400 WWW: http://www.RNI.Helsinki.FI/~boh/ Journal of Negative Results - EEB: www.jnr-eeb.org __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Storing vectors as vectors and iterating through them
Ken -- try using lists: > vec.list <- list(1:5, 2:4, 3:8) > vec.list <- c(vec.list, list(7:9)) > vec.list [[1]] [1] 1 2 3 4 5 [[2]] [1] 2 3 4 [[3]] [1] 3 4 5 6 7 8 [[4]] [1] 7 8 9 Then you can use "lapply" or "sapply", or just a "for" loop, to iterate over the list, applying your function to each element. Hope this helps, Matt Wiener -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Ken Termiso Sent: Monday, April 18, 2005 1:09 PM To: r-help@stat.math.ethz.ch Subject: [R] Storing vectors as vectors and iterating through them Hi all, I have a bunch of int vectors. Each vector holds a bunch of ints that correspond to row numbers of an existing matrix. I use the int vectors to pull out rows of data from a matrix, i.e. data <- my_matrix[int_vector,] I would like to store these int vectors in some sort of data structure that will preserve them as-is and allow iteration. I guess what I'm looking for would be something analogous to the java Vector class, as in this java-like pseudocode : Vector V = new Vector; V.add(a,b,c) // where a,b,c are lists for(int i = 0; ihttps://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Storing vectors as vectors and iterating through them
Hi all, I have a bunch of int vectors. Each vector holds a bunch of ints that correspond to row numbers of an existing matrix. I use the int vectors to pull out rows of data from a matrix, i.e. data <- my_matrix[int_vector,] I would like to store these int vectors in some sort of data structure that will preserve them as-is and allow iteration. I guess what I'm looking for would be something analogous to the java Vector class, as in this java-like pseudocode : Vector V = new Vector; V.add(a,b,c) // where a,b,c are lists for(int i = 0; i The point is to iterate through the data structure containing the int vectors, and, for each int vector, do some clustering and plotting, but what I cannot find is a data structure in R that would support this.. trying c(a,b,c) does not preserve each int vector, but instead merges all the ints into one vector. I need to keep them separate (so I can create a separate plot for each vector). Thanks in advance, Ken __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] refitting lm() with same x, different y
William, As a first shot, use lm with a matrix response. That fits them all at once with one QR-decomposition. No analogue for glm or lmer, though, since for those the iterative fits run do depend on the response. Brian On Mon, 18 Apr 2005, William Valdar wrote: Dear All, Is there is a fast way of refitting lm() when the design matrix stays constant but the response is different? For example, y1 ~ X y2 ~ X y3 ~ X ...etc. where y1 is the 1st instance of the response vector. Calling lm() every time seems rather wasteful since the QR-decomposition of X needs to be calculated only once. It would be nice if qr() was called only once and then the same QR-factorization used in all subsequent fits. However, I can't see a way to do this easily. Can anybody else? Why do I want to do this? I'm fitting ~1000 different X's to a response vector (for biologists: 1000 genetic markers to a measured phenotype with 2000 cases) and wish to establish global significance thresholds for multiple testing. The fits have a complex dependency structure that makes the Bonferroni correction inappropriate. So I intend to refit all ~1000 X's with a shuffled response many times. However, this runs too slow for my needs. Of course, not having to redo QR will only help if QR is a rate limiting step in lm(), so if anybody can tell me it's not, then that would be very helpful too. I would also like to do this for glm() and lmer() fits. Ideally. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Very Slow Gower Similarity Function
On 18 Apr 2005, at 19:10, Tyler Smith wrote: Hello, I am a relatively new user of R. I have written a basic function to calculate the Gower similarity function. I was motivated to do so partly as an excercise in learning R, and partly because the existing option (vegdist in the vegan package) does not accept missing values. Speed is the reason to use C instead of R. It should be easy, almost trivial, to modify the vegdist.c so that it handles missing values. I guess this handling means ignoring the value pair if one of the values is missing -- which is not so gentle to the metric properties so dear to Gower. Package vegan is designed for ecological community data which generally do not have missing values (except in environmental data), but contributions are welcome. I think I have succeeded - my function gives me the correct values. However, now that I'm starting to use it with real data, I realise it's very slow. It takes more than 45 minutes on my Windows 98 machine (R 2.0.1 Patched (2005-03-29)) with a 185x32 matrix with ca 100 missing values. If anyone can suggest ways to speed up my function I would appreciate it. I suspect having a pair of nested for loops is the problem, but I couldn't figure out how to get rid of them. cheers, jari oksanen -- Jari Oksanen, Oulu, Finland __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] colClasses = "Date" in read.delim, how to pass date-format?
You are confusing class "Date" (part of R) with class "dates" (part of
package chron). There is no as() method for class "dates", so you can't
do this. You can read the column as character (not factor) and convert
later, but it sounds like the `huge amount of memory (and time)' is in
fact taken by package chron.
On Mon, 18 Apr 2005, Christoph Lehmann wrote:
I have a huge data-set with one column being of type date.
Of course I can import the data using this column as "factor" and then
convert it later to dates, using:
sws.bezuege$FaktDat <- dates(as.character(sws.bezuege$FaktDat),
format = c(dates = "d.m.y"))
But the conversion requires a huge amount of memory (and time), therefore I
would like to use colClasses = c("Date"). My question is:
since I have format = c(dates = "d.m.y"), how can I pass this option to
read.delim(..., colClasses = c("Date")) ?
--
Brian D. Ripley, [EMAIL PROTECTED]
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UKFax: +44 1865 272595
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] a small problem with data
Yoko Nakajima wrote: Hello, I have a strange problem. I may have specified and scanned the data set (I receive a message: Read 384134 items). I can check the dimension of the data set by dim(xyz). But when I try to load the data set or to list the data set by data(xyz), then I receive an error message saying that data set not found in: data(xyz). Is this possible, or is this because I deal with a large data set (dimension is 13246 x 29). I can not run any package with this data set. I am guessing that this is because of the same reason that I can not load/list the data set. Are reading data and loading data are different? I am very puzzled. The data function is used to load data sets from a package (and is not even needed for that any more if the package uses LazyData). You do not need to use data(xyz) before using the name xyz. To check the names of the objects currently in your worksheet use objects() __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Latent Class Analyisis in R
Helen, You may consider using LCA 1.1, written by Niels Waller at Vanderbilt University. You may need to get it from his website at Vanderbilt instead of CRAN, at least it was the case in the past. Tim Liao Original message >Date: Mon, 18 Apr 2005 12:29:41 -0400 >From: "Dennis, Helen" <[EMAIL PROTECTED]> >Subject: [R] Latent Class Analyisis in R >To: > >Hi, > >I am new to R. I would like to know if there is a package available that >will estimate a 2-class Latent Class Analysis, with trichotomous >observed variables, some of which are free to covary. > >I have read the PDF files regarding the e1071 and gllm packages, but it >appears that these only allow dichotomous measured variables, and it is >not clear to me whether they allow the measured variables to covary. > >Thanks! > >Helen Dennis >Education Associate >Assessment and Analysis Group >Delaware Dept. of Education >401 Federal Street >Suite 2 >Dover, DE 19707 >[EMAIL PROTECTED] >tel: 302-739-6700 >fax: 302-739-3092 > >__ >R-help@stat.math.ethz.ch mailing list >https://stat.ethz.ch/mailman/listinfo/r-help >PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Retrieving column descriptions into Stata
Dear All,
I am working on a project with Stata 7 datafiles. I'm trying to do some
of my analysis in R since I don't have Stata handy. My problem is with
accessing the Stata column descriptions. (Part of the project is to
clean up these column descriptions.)
I imported the Stata files ok using the standard
X <- read.dta("c:/testdata/test1.dta")
test1.dta has column descriptions, but I can't figure out where they are
kept in X. (I assume they are in X somewhere; at least this is what the
docs seem to imply.) I see variable names in X, but no descriptions.
As a reverse test, I created a short dataset in R (the numbers 1-100),
exported it to Stata format, and opened it in Stata. I was expecting
the column description to be null, but the column description was
populated with the variable name. So it appears that the export program
knows how to set column descriptions. I just need to figure out what
attributes I need to check/set to make it do so.
Any suggestions would be much appreciated.
Regards,
AJ Bostian
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Latent Class Analyisis in R
Hi, I am new to R. I would like to know if there is a package available that will estimate a 2-class Latent Class Analysis, with trichotomous observed variables, some of which are free to covary. I have read the PDF files regarding the e1071 and gllm packages, but it appears that these only allow dichotomous measured variables, and it is not clear to me whether they allow the measured variables to covary. Thanks! Helen Dennis Education Associate Assessment and Analysis Group Delaware Dept. of Education 401 Federal Street Suite 2 Dover, DE 19707 [EMAIL PROTECTED] tel: 302-739-6700 fax: 302-739-3092 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Creating packages with windows (accessing data)
Thank you, The DESCRIPTION is what I needed...all works as I want now. Alan- Alan T. Arnholt Associate Professor Department of Mathematical Sciences Appalachian State University Boone, NC 28608 (828) 262 2863 www1.appstate.edu/~arnholta -Original Message- From: Prof Brian Ripley [mailto:[EMAIL PROTECTED] Sent: Sunday, April 17, 2005 10:24 AM To: Alan Arnholt Cc: r-help@stat.math.ethz.ch; [EMAIL PROTECTED] Subject: Re: [R] Creating packages with windows (accessing data) Look at `Writing R Extensions' and the description of the DESCRIPTION file, specifically `LazyData'. That is the manual about packages On Sun, 17 Apr 2005, Alan Arnholt wrote: > I have created a package (under Windows 2.0.1) with 300+ data sets and <20 > or so functions I use in teaching. However, to access the data, one needs > to type data(foo) once the package has been installed and loaded. With > other packages namely MASS, after the package is installed and loaded with > library(MASS), it is possible to refer to a data set say Animals by simply > typing Animals at the command prompt. I would like to have similar > functionality in my package. Would someone provide some hints as to what > I need to do (read about xxx...or provide a line of code) so that the data > in the package can be accessed once it is loaded without typing data(foo) > all the time. Currently, when I type the name of a function it shows the > code for the function. I would like it to also show the data when I type > the name of any of the data sets. Thanks for the pointers in advance. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Very Slow Gower Similarity Function
Hello,
I am a relatively new user of R. I have written a basic function to calculate
the Gower similarity function. I was motivated to do so partly as an excercise
in learning R, and partly because the existing option (vegdist in the vegan
package) does not accept missing values.
I think I have succeeded - my function gives me the correct values. However, now
that I'm starting to use it with real data, I realise it's very slow. It takes
more than 45 minutes on my Windows 98 machine (R 2.0.1 Patched (2005-03-29))
with a 185x32 matrix with ca 100 missing values. If anyone can suggest ways to
speed up my function I would appreciate it. I suspect having a pair of nested
for loops is the problem, but I couldn't figure out how to get rid of them.
The function is:
### Gower Similarity Matrix###
sGow <- function (mat){
OBJ <- nrow(mat) #number of objects
MATDESC <- ncol (mat) #number of descriptors
MRANGE <- apply (mat,2,max, na.rm=T)-apply (mat,2,min,na.rm=T) #descr ranges
DESCRIPT <- 1:MATDESC #descriptor index vector
smat <- matrix(1, nrow = OBJ, ncol = OBJ) #'empty' similarity matrix
for (i in 1:OBJ){
for (j in i:OBJ){
##calculate index vector of non-NA descriptors between objects i and j
descvect <- intersect (setdiff (DESCRIPT, DESCRIPT[is.na(mat[i,DESCRIPT])]),
setdiff (DESCRIPT, DESCRIPT[is.na (mat[j,DESCRIPT])]))
descnum <- length(descvect) # number of valid descr for i~j comparison
partialsim <- (1- abs(mat[i,descvect]-mat[j,descvect])/MRANGE[descvect])
smat[i,j] <- smat[j,i] <- sum (partialsim) / descnum
}
}
smat
}
Thank-you for your time,
Tyler
--
Tyler Smith
PhD Candidate
Plant Science Department
McGill University
[EMAIL PROTECTED]
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Barplot and colors for legend
>>>Hi all! >>> >>>One quick question: How do I get the standard colors >>>used by barplot? I have >>>some stacked bars and would like to add a horizontal >>>legend via legend() but I >>>don't know how to find the colors for the fills of the >>>legend points. >>> >>>Thanks! >>> Werner >> >> >>Type >> barplot.default >>and read the code: >> >>for a vector: "grey", >>for a matrix: grey(seq(0.3^2.2, 0.9^2.2, length = >>nrow(height))^(1/2.2)) > > > This must be an old version of R ;-) > > In R 2.1.0, there is a function gray.colors() which creates a vector of > gamma-corrected gray colors (and is used in barplot.default). > > Best, > Z Thanks for the quick help! I am using R 2.0.1 under Windows 2000 and Uwes suggestion helps. And another thanks: Finally I realized that I don't have the newest R version! Thanks again, Werner __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Install problem on Solaris 9
I'm trying to install R-2.0.1 on Solaris 9 and I'm receiving the following error messages during make. begin installing recommended package foreign make[2]: *** [foreign.ts] Error 1 make[2]: Leaving directory `/opt/net/source/R-2.0.1/src/library/Recommended' make[1]: *** [recommended-packages] Error 2 make[1]: Leaving directory `/opt/net/source/R-2.0.1/src/library/Recommended' make: *** [stamp-recommended] Error 2 A review of src/library/Recommended/foreign.ts.out shows: * Installing *source* package 'foreign' ... configure: loading cache /dev/null checking for gcc... make[3]: Entering directory `/tmp/R.INSTALL.169/foreign' gcc -m64 make[3]: Leaving directory `/tmp/R.INSTALL.169/foreign' checking for C compiler default output file name... configure: error: C compiler cannot create executables See `config.log' for more details. ERROR: configuration failed for package 'foreign' A review of config.log does not reveal anything terribly useful. I'm using gcc, g77, and g++ v3.4.1, 32 bit builds (I think), all installed under /opt/net/utils/gcc3.4.1, which is in the path. LD_LIBRARY_PATH=/otherstuff:/opt/net/utils/gcc3.4.1/lib:/opt/net/utils/l ib:/otherstuff My config.site contains: CC="gcc" CPPFLAGS="-I/usr/include" F77="g77" LDFLAGS="-L/opt/net/utils/gcc3.4.1/lib/" CXX="g++" What should I be looking for here? Thanks. Rusty [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] a small problem with data
Hello, I have a strange problem. I may have specified and scanned the data set (I receive a message: Read 384134 items). I can check the dimension of the data set by dim(xyz). But when I try to load the data set or to list the data set by data(xyz), then I receive an error message saying that data set not found in: data(xyz). Is this possible, or is this because I deal with a large data set (dimension is 13246 x 29). I can not run any package with this data set. I am guessing that this is because of the same reason that I can not load/list the data set. Are reading data and loading data are different? I am very puzzled. Sincerely, Yoko. [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] R2.0.1 for Mac OS X 10.3 problem
On Mon, 2005-04-18 at 06:39 -0700, Horacio Montenegro wrote: > I have the same problem, in Windows, and I think > the .Rdata is not corrupted. Load R without loading > the.Rdata file - move or rename it. Then library(lme4) > and load the .Rdata - it should work. > This also is my experience (in Linux & MacOS X). The .RData need not be corrupted, but you have "corrupted" your R installation by deleting or corrupting some package using S4 methods (like failed upgrade of a S4 method package). When you start R so that it tries to restore those S4 objects in .RData, you get the error. Renaming or deleting .RData will help, of course. Alternatively, in my case it helped to start R with option --no-restore-data (in Mac when starting R from terminal -- where you are all the time in Linux). Probably it would help to install again the original S4 package. (In my case this happened when I tried Thomas Yee's VGAM and then removed the package.) cheers, jari oksanen -- Jari Oksanen -- Dept Biology, Univ Oulu, 90014 Oulu, Finland email [EMAIL PROTECTED], homepage http://cc.oulu.fi/~jarioksa/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] refitting lm() with same x, different y
Dear All, Is there is a fast way of refitting lm() when the design matrix stays constant but the response is different? For example, y1 ~ X y2 ~ X y3 ~ X ...etc. where y1 is the 1st instance of the response vector. Calling lm() every time seems rather wasteful since the QR-decomposition of X needs to be calculated only once. It would be nice if qr() was called only once and then the same QR-factorization used in all subsequent fits. However, I can't see a way to do this easily. Can anybody else? Why do I want to do this? I'm fitting ~1000 different X's to a response vector (for biologists: 1000 genetic markers to a measured phenotype with 2000 cases) and wish to establish global significance thresholds for multiple testing. The fits have a complex dependency structure that makes the Bonferroni correction inappropriate. So I intend to refit all ~1000 X's with a shuffled response many times. However, this runs too slow for my needs. Of course, not having to redo QR will only help if QR is a rate limiting step in lm(), so if anybody can tell me it's not, then that would be very helpful too. I would also like to do this for glm() and lmer() fits. Ideally. Many thanks, William =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Dr William Valdar ++44 (0)1865 287 717 Wellcome Trust Centre [EMAIL PROTECTED] for Human Genetics, Oxford www.well.ox.ac.uk/~valdar __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] colClasses = "Date" in read.delim, how to pass date-format?
On 4/18/05, Christoph Lehmann <[EMAIL PROTECTED]> wrote:
> Hi
> I have a huge data-set with one column being of type date.
> Of course I can import the data using this column as "factor" and then
> convert it later to dates, using:
>
> sws.bezuege$FaktDat <- dates(as.character(sws.bezuege$FaktDat),
> format = c(dates = "d.m.y"))
>
> But the conversion requires a huge amount of memory (and time),
> therefore I would like to use colClasses = c("Date"). My question is:
> since I have format = c(dates = "d.m.y"), how can I pass this option to
> read.delim(..., colClasses = c("Date")) ?
>
Check out:
http://tolstoy.newcastle.edu.au/R/help/05/02/12003.html
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] the graph gallery strikes back
Gabor, I was obviously kidding. Do you really think I would build a gallery for R without knowing its website. I spend a lot of time around http://www.r-project.org/ The link has been added on the minute after i did read your mail. Romain. Le 18.04.2005 16:54, Gabor Grothendieck a écrit : Its the R home page. I think the graph gallery might be used to direct people to R before they have used it since its quite attractive. At that point they will want to know how to get to the rest of R. On 4/18/05, Romain Francois <[EMAIL PROTECTED]> wrote: Hi, I don't know about that web page, what is it ? Romain Le 18.04.2005 16:33, Gabor Grothendieck a écrit : Just one suggestion. Maybe you could provide on a link on that page to http://www.r-project.org . On 4/18/05, Gabor Grothendieck <[EMAIL PROTECTED]> wrote: Yes, it works for me now. Thanks. Very impressive I might add. On 4/18/05, Romain Francois <[EMAIL PROTECTED]> wrote: Oops :) That should be allright now. Romain Le 18.04.2005 13:52, Naji a écrit : Hi Romain, Great job.. Thanks a lot. When pointing at the graph or the link (ie Kernel density estimator in R², Perspective plot and contour plot, etc) the browser is sending back a SQL error (Safari, MacOSX3.8) Erreur SQL! Unknown column 'qualification' in 'field list' Best regards Naji Le 18/04/05 11:29, « Romain Francois » <[EMAIL PROTECTED]> a écrit : Hello useRs and helpRs, Some time ago, in a gallaxy far away (here is the thread : http://finzi.psych.upenn.edu/R/Rhelp02a/archive/46532.html ) we discussed about a graph gallery showing the power of R in that domain. I did some work around that, and there is a (pretty advanced) draft here : http://addictedtor.free.fr/graphiques/displayGallery.php For instance, there are some of my graphs, some of Eric Lecoutre's and some coming from demo(graphics), demo(image), demo(persp) and so on. Pretty soon, I'll add the possibility to : - add graphs dynamically from the web - give a mark to each graph (maybe we can do some stats :) ) - link R function to their help pages from http://finzi.psych.upenn.edu/R - add references to publications - whatever Please take a look and tell me what should be improved. May the foRce be with you. Romain. PS1 : Regarding graphs from demo(*), I wrote that author is R Development Core Team, maybe those were done by a precise person in the Core, in that case tell me and I'll change it. PS2 : Source code highlighting was done by the highlight software : ( http://www.andre-simon.de/ ) suggested to me by Eric Lecoutre. -- ~ ~~ Romain FRANCOIS - http://addictedtor.free.fr ~~ Etudiant ISUP - CS3 - Industrie et Services ~~http://www.isup.cicrp.jussieu.fr/ ~~ Stagiaire INRIA Futurs - Equipe SELECT ~~ http://www.inria.fr/recherche/equipes/select.fr.html~~ ~ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] generalized linear mixed models - how to compare?
No puedo entender. Nicht versteh. Je ne comprend pas.
Peter Spreeuwenberg wrote:
Liset
Dat moet lukken en hoe sneller je het anlevert hoe eerder je resultaten terug
ziet.
Dus, Ik wacht af.
groet Peter S
Date sent: Sun, 17 Apr 2005 18:07:28 +0100 (BST)
From: Prof Brian Ripley <[EMAIL PROTECTED]>
To: Deepayan Sarkar <[EMAIL PROTECTED]>
Subject:Re: [R] generalized linear mixed models - how to
compare?
Copies to: r-help@stat.math.ethz.ch,
Nestor Fernandez <[EMAIL PROTECTED]>
On Sun, 17 Apr 2005, Deepayan Sarkar wrote:
On Sunday 17 April 2005 08:39, Nestor Fernandez wrote:
I want to evaluate several generalized linear mixed models, including
the null model, and select the best approximating one. I have tried
glmmPQL (MASS library) and GLMM (lme4) to fit the models. Both result
in similar parameter estimates but fairly different likelihood
estimates.
My questions:
1- Is it correct to calculate AIC for comparing my models, given that
they use quasi-likelihood estimates? If not, how can I compare them?
2- Why the large differences in likelihood estimates between the two
procedures?
The likelihood reported by glmmPQL is wrong, as it's the likelihood of
an incorrect model (namely, an lme model that approximates the correct
glmm model).
Actually glmmPQL does not report a likelihood. It returns an object of
class "lme", but you need to refer to the reference for how to interpret
that. It *is* support software for a book.
GLMM uses (mostly) the same procedure to get parameter estimates, but as
a final step calculates the likelihood for the correct model for those
estimates (so the likelihood reported by it should be fairly reliable).
Well, perhaps but I need more convincing. The likelihood involves many
high-dimensional non-analytic integrations, so I do not see how GLMM can
do those integrals -- it might approximate them, but that would not be
`calculates the likelihood for the correct model'. It would be helpful to
have a clarification of this claim. (Our experiments show that finding an
accurate value of the log-likelihood is difficult and many available
pieces of software differ in their values by large amounts.)
Further, since neither procedure does ML fitting, this is not a maximized
likelihood as required to calculate an AIC value. And even if it were,
you need to be careful as often one GLMM is a boundary value for another,
in which case the theory behind AIC needs adjustment.
--
Brian D. Ripley, [EMAIL PROTECTED]
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UKFax: +44 1865 272595
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Stichting NIVEL / NETHERLANDS INSTITUTE OF PRIMARY HEALTH CARE
Peter Spreeuwenberg
P.O. Box 1568
3500 BN Utrecht
Netherland
E-mail : [EMAIL PROTECTED]
Direct: +31-30-2729678
General : +31-30-2729700
Fax : +31-30-2729729
Web-site http://www.nivel.nl
DISCLAIMER__
This message contains information that may be privileged or\...{{dropped}}
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] colClasses = "Date" in read.delim, how to pass date-format?
Hi
I have a huge data-set with one column being of type date.
Of course I can import the data using this column as "factor" and then
convert it later to dates, using:
sws.bezuege$FaktDat <- dates(as.character(sws.bezuege$FaktDat),
format = c(dates = "d.m.y"))
But the conversion requires a huge amount of memory (and time),
therefore I would like to use colClasses = c("Date"). My question is:
since I have format = c(dates = "d.m.y"), how can I pass this option to
read.delim(..., colClasses = c("Date")) ?
thanks for a hint
cheers
christoph
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Barplot and colors for legend
On Mon, 18 Apr 2005 16:48:42 +0200 Uwe Ligges wrote: > Werner Wernersen wrote: > > > Hi all! > > > > One quick question: How do I get the standard colors > > used by barplot? I have > > some stacked bars and would like to add a horizontal > > legend via legend() but I > > don't know how to find the colors for the fills of the > > legend points. > > > > Thanks! > >Werner > > > Type >barplot.default > and read the code: > > for a vector: "grey", > for a matrix: grey(seq(0.3^2.2, 0.9^2.2, length = > nrow(height))^(1/2.2)) This must be an old version of R ;-) In R 2.1.0, there is a function gray.colors() which creates a vector of gamma-corrected gray colors (and is used in barplot.default). Best, Z > > Uwe Ligges > > __ > R-help@stat.math.ethz.ch mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! > http://www.R-project.org/posting-guide.html > __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] the graph gallery strikes back
Its the R home page. I think the graph gallery might be used to direct people to R before they have used it since its quite attractive. At that point they will want to know how to get to the rest of R. On 4/18/05, Romain Francois <[EMAIL PROTECTED]> wrote: > Hi, > > I don't know about that web page, what is it ? > > Romain > > Le 18.04.2005 16:33, Gabor Grothendieck a écrit : > > >Just one suggestion. Maybe you could provide on a link on that > >page to http://www.r-project.org . > > > >On 4/18/05, Gabor Grothendieck <[EMAIL PROTECTED]> wrote: > > > > > >>Yes, it works for me now. Thanks. Very impressive I might add. > >> > >>On 4/18/05, Romain Francois <[EMAIL PROTECTED]> wrote: > >> > >> > >>>Oops :) > >>> > >>>That should be allright now. > >>> > >>>Romain > >>> > >>>Le 18.04.2005 13:52, Naji a écrit : > >>> > >>> > >>> > Hi Romain, > > > Great job.. Thanks a lot. > When pointing at the graph or the link (ie Kernel density estimator in R², > Perspective plot and contour plot, etc) the browser is sending back a SQL > error (Safari, MacOSX3.8) > > Erreur SQL! > Unknown column 'qualification' in 'field list' > > Best regards > Naji > > > Le 18/04/05 11:29, « Romain Francois » <[EMAIL PROTECTED]> a écrit : > > > > > > >Hello useRs and helpRs, > > > >Some time ago, in a gallaxy far away (here is the thread : > >http://finzi.psych.upenn.edu/R/Rhelp02a/archive/46532.html ) we > >discussed about a graph gallery showing the power of R in that domain. I > >did some work around that, and there is a (pretty advanced) draft here : > > > >http://addictedtor.free.fr/graphiques/displayGallery.php > > > >For instance, there are some of my graphs, some of Eric Lecoutre's and > >some coming from demo(graphics), demo(image), demo(persp) and so on. > > > >Pretty soon, I'll add the possibility to : > >- add graphs dynamically from the web > >- give a mark to each graph (maybe we can do some stats :) ) > >- link R function to their help pages from http://finzi.psych.upenn.edu/R > >- add references to publications > >- whatever > > > >Please take a look and tell me what should be improved. May the foRce be > >with you. > > > >Romain. > > > >PS1 : Regarding graphs from demo(*), I wrote that author is R > >Development Core Team, maybe those were done by a precise person in the > >Core, in that case tell me and I'll change it. > > > >PS2 : Source code highlighting was done by the highlight software : ( > >http://www.andre-simon.de/ ) suggested to me by Eric Lecoutre. > > > > > > > > > > > > > > > > -- > ~ > ~~ Romain FRANCOIS - http://addictedtor.free.fr ~~ > Etudiant ISUP - CS3 - Industrie et Services > ~~http://www.isup.cicrp.jussieu.fr/ ~~ > Stagiaire INRIA Futurs - Equipe SELECT > ~~ http://www.inria.fr/recherche/equipes/select.fr.html~~ > ~ > > __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] the graph gallery strikes back
Hi, I don't know about that web page, what is it ? Romain Le 18.04.2005 16:33, Gabor Grothendieck a écrit : Just one suggestion. Maybe you could provide on a link on that page to http://www.r-project.org . On 4/18/05, Gabor Grothendieck <[EMAIL PROTECTED]> wrote: Yes, it works for me now. Thanks. Very impressive I might add. On 4/18/05, Romain Francois <[EMAIL PROTECTED]> wrote: Oops :) That should be allright now. Romain Le 18.04.2005 13:52, Naji a écrit : Hi Romain, Great job.. Thanks a lot. When pointing at the graph or the link (ie Kernel density estimator in R², Perspective plot and contour plot, etc) the browser is sending back a SQL error (Safari, MacOSX3.8) Erreur SQL! Unknown column 'qualification' in 'field list' Best regards Naji Le 18/04/05 11:29, « Romain Francois » <[EMAIL PROTECTED]> a écrit : Hello useRs and helpRs, Some time ago, in a gallaxy far away (here is the thread : http://finzi.psych.upenn.edu/R/Rhelp02a/archive/46532.html ) we discussed about a graph gallery showing the power of R in that domain. I did some work around that, and there is a (pretty advanced) draft here : http://addictedtor.free.fr/graphiques/displayGallery.php For instance, there are some of my graphs, some of Eric Lecoutre's and some coming from demo(graphics), demo(image), demo(persp) and so on. Pretty soon, I'll add the possibility to : - add graphs dynamically from the web - give a mark to each graph (maybe we can do some stats :) ) - link R function to their help pages from http://finzi.psych.upenn.edu/R - add references to publications - whatever Please take a look and tell me what should be improved. May the foRce be with you. Romain. PS1 : Regarding graphs from demo(*), I wrote that author is R Development Core Team, maybe those were done by a precise person in the Core, in that case tell me and I'll change it. PS2 : Source code highlighting was done by the highlight software : ( http://www.andre-simon.de/ ) suggested to me by Eric Lecoutre. -- ~ ~~ Romain FRANCOIS - http://addictedtor.free.fr ~~ Etudiant ISUP - CS3 - Industrie et Services ~~http://www.isup.cicrp.jussieu.fr/ ~~ Stagiaire INRIA Futurs - Equipe SELECT ~~ http://www.inria.fr/recherche/equipes/select.fr.html~~ ~ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Barplot and colors for legend
Werner Wernersen wrote: Hi all! One quick question: How do I get the standard colors used by barplot? I have some stacked bars and would like to add a horizontal legend via legend() but I don't know how to find the colors for the fills of the legend points. Thanks! Werner Type barplot.default and read the code: for a vector: "grey", for a matrix: grey(seq(0.3^2.2, 0.9^2.2, length = nrow(height))^(1/2.2)) Uwe Ligges __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Barplot and colors for legend
Hi all! One quick question: How do I get the standard colors used by barplot? I have some stacked bars and would like to add a horizontal legend via legend() but I don't know how to find the colors for the fills of the legend points. Thanks! Werner __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] the graph gallery strikes back
Just one suggestion. Maybe you could provide on a link on that page to http://www.r-project.org . On 4/18/05, Gabor Grothendieck <[EMAIL PROTECTED]> wrote: > Yes, it works for me now. Thanks. Very impressive I might add. > > On 4/18/05, Romain Francois <[EMAIL PROTECTED]> wrote: > > Oops :) > > > > That should be allright now. > > > > Romain > > > > Le 18.04.2005 13:52, Naji a écrit : > > > > >Hi Romain, > > > > > > > > >Great job.. Thanks a lot. > > >When pointing at the graph or the link (ie Kernel density estimator in R², > > >Perspective plot and contour plot, etc) the browser is sending back a SQL > > >error (Safari, MacOSX3.8) > > > > > >Erreur SQL! > > >Unknown column 'qualification' in 'field list' > > > > > >Best regards > > >Naji > > > > > > > > >Le 18/04/05 11:29, « Romain Francois » <[EMAIL PROTECTED]> a écrit : > > > > > > > > > > > >>Hello useRs and helpRs, > > >> > > >>Some time ago, in a gallaxy far away (here is the thread : > > >>http://finzi.psych.upenn.edu/R/Rhelp02a/archive/46532.html ) we > > >>discussed about a graph gallery showing the power of R in that domain. I > > >>did some work around that, and there is a (pretty advanced) draft here : > > >> > > >>http://addictedtor.free.fr/graphiques/displayGallery.php > > >> > > >>For instance, there are some of my graphs, some of Eric Lecoutre's and > > >>some coming from demo(graphics), demo(image), demo(persp) and so on. > > >> > > >>Pretty soon, I'll add the possibility to : > > >>- add graphs dynamically from the web > > >>- give a mark to each graph (maybe we can do some stats :) ) > > >>- link R function to their help pages from http://finzi.psych.upenn.edu/R > > >>- add references to publications > > >>- whatever > > >> > > >>Please take a look and tell me what should be improved. May the foRce be > > >>with you. > > >> > > >>Romain. > > >> > > >>PS1 : Regarding graphs from demo(*), I wrote that author is R > > >>Development Core Team, maybe those were done by a precise person in the > > >>Core, in that case tell me and I'll change it. > > >> > > >>PS2 : Source code highlighting was done by the highlight software : ( > > >>http://www.andre-simon.de/ ) suggested to me by Eric Lecoutre. > > >> > > >> > > >> > > > > > > > > > > > > > > > > > > > > > > -- > > ~ > > ~~ Romain FRANCOIS - http://addictedtor.free.fr ~~ > > Etudiant ISUP - CS3 - Industrie et Services > > ~~http://www.isup.cicrp.jussieu.fr/ ~~ > > Stagiaire INRIA Futurs - Equipe SELECT > > ~~ http://www.inria.fr/recherche/equipes/select.fr.html~~ > > ~ > > > > __ > > R-help@stat.math.ethz.ch mailing list > > https://stat.ethz.ch/mailman/listinfo/r-help > > PLEASE do read the posting guide! > > http://www.R-project.org/posting-guide.html > > > __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] nls error in formula
thank you, it worked. Sylvia --- Douglas Bates <[EMAIL PROTECTED]> wrote: > Anaid Diaz wrote: > > Hi, > > I'm a new R user, with a lot of questions. At the > > moment I'm stoped on an error traying to fit a > model: > > > > > >>x <- sandeel ## numeric data (2500-6) > >>y <- Noss ## numeric data (0-1.2) > >>A <- 0.8 > >>B <- 0.6 > >>C <- 1/4 > >>nls( y ~ A-B*exp(-C*x)) > > > > Error in match.call(definition, call, expand.dots) > : > > .Primitive... is not a function > > > > I'm not sure if the error is due to the formula > sytax > > or something else. > > I'll appreciate any help > > > > Sylvia > > Use > > nls(y ~ A -B*exp(-C*x), start = c(A = 0.8, B = 0.6, > C = 1/4)) > > instead or, preferably, > > nls(y ~ SSasymp(x, Asym, R0, lrc)) > > and read > > ?SSasymp > > for an interpretation of the estimated parameters. > _ Do You Yahoo!? La mejor conexión a internet y 25MB extra a tu correo por $100 al mes. http://net.yahoo.com.mx __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] nls error in formula
Anaid Diaz wrote: Hi, I'm a new R user, with a lot of questions. At the moment I'm stoped on an error traying to fit a model: x <- sandeel ## numeric data (2500-6) y <- Noss ## numeric data (0-1.2) A <- 0.8 B <- 0.6 C <- 1/4 nls( y ~ A-B*exp(-C*x)) Error in match.call(definition, call, expand.dots) : .Primitive... is not a function I'm not sure if the error is due to the formula sytax or something else. I'll appreciate any help Sylvia Use nls(y ~ A -B*exp(-C*x), start = c(A = 0.8, B = 0.6, C = 1/4)) instead or, preferably, nls(y ~ SSasymp(x, Asym, R0, lrc)) and read ?SSasymp for an interpretation of the estimated parameters. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] the graph gallery strikes back
Romain Francois <[EMAIL PROTECTED]> writes: > Oops :) > > That should be allright now. > > Romain Yep. You might want to antialias the small images though. And, btw, just for fun you might want to include the real Maunga Whau among the volcano pictures: http://www.biostat.ku.dk/~pd/auckland2000/maungawhau.jpg You will notice that the real world is actually closer to pictures 25 and 26, but 27 looks more dangerous and volcano-like... -- O__ Peter Dalgaard Blegdamsvej 3 c/ /'_ --- Dept. of Biostatistics 2200 Cph. N (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Generating a binomial random variable correlated with a
You may generate a single standard normal random variable Z and set
X = (Z>x). According to back-of-the-envelope calculations, the two
have a correlation of
exp(-x^2/2)/sqrt(2*pi*pnorm(x)*(1-pnorm(x)))
which goes from a maximum of about 0.79 at x=0 to 0 for x going to
infinity.
If you aim at a correlation of 0.7, this is how to go:
> rt <- uniroot(function(x) exp(-x^2/2)/sqrt(2*pi*pnorm(x)*(1-pnorm(x)))-0.7,
+ lower=0,upper=3)
> rt$root
[1] 0.841188
> x <- rnorm(1000)
> y <- as.numeric(x>rt$root)
> cor(x,y)
[1] 0.7069753
Best,
Giovanni
--
__
[ ]
[ Giovanni Petris [EMAIL PROTECTED] ]
[ Department of Mathematical Sciences ]
[ University of Arkansas - Fayetteville, AR 72701 ]
[ Ph: (479) 575-6324, 575-8630 (fax) ]
[ http://definetti.uark.edu/~gpetris/ ]
[__]
> Date: Sun, 17 Apr 2005 11:52:49 +0100 (BST)
> From: [EMAIL PROTECTED] (Ted Harding)
> Sender: [EMAIL PROTECTED]
> Cc: r-help@stat.math.ethz.ch
> Precedence: list
>
> On 16-Apr-05 Ashraf Chaudhary wrote:
> > Ted:
> > Thank you for your help. All I want is a binomial random
> > variable that is correlated with a normal random variable
> > with specified correlation. By linear I mean the ordinary
> > Pearson correlation. I tried the following two methods,
> > in each case the resulting correlation is substantially
> > less than the one specified.
> >
> > Method I: Generate two correlated normals (using Cholesky
> > decomposition method) and dichotomize one (0/1) to get the
> > binomial. Method II: Generate two correlated variables, one
> > binomial and one normal using the Cholesky decomposition methods.
> >
> > Here is how I did:
> >
> > X <- rnorm(100)
> > Y <- rnorm(100)
> > r<- 0.7
> > Y1 <- X*r+Y*sqrt(1-r**2)
> > cor(X,Y1) # Correlated normals using Cholesky decomposition
> > cor(X>0.84,Y1) # Method I
> >
> >##
> > X1 <- rbinom(100,1,0.5)
> > Y2 <- X1*r+Y*sqrt(1-r**2)
> > cor(X1,Y2); # Method II
>
> Hello Ashraf,
>
> The above more explicit explanation certainly helps to clarify
> your question! (I echo Bill's counsel to spell out essential
> detail).
>
> I'll lead on from your "second method" above, which makes it
> explicit that, to each of your 100 values (Y) sampled from rnorm
> you are sampling from {0,1} to get one of your 100 binomial
> (n=1) variables (X1). However, in this second method you also
> create the derived variable Y2 = X1*r+Y*sqrt(1-r**2), and
> apparently you want this (Y2) to be the Normal variate which
> is correlated with X1.
>
> Unfortunately, given the way Y2 is constructed, it does not
> have a Normal distribution. This is almost obvious from the
> fact that, conditional on the number of 1s in X1, the values
> of Y2 are a mixture of N(0,1-r^2) and N(r,1-r^2).
>
> You can explore this in R by looking at the histogram of a
> much larger sample of Y2. Thus:
>
> r<- 0.7
> Y <- rnorm(10)
> X1 <- rbinom(10,1,0.5)
> Y2 <- X1*r+Y*sqrt(1-r**2)
> m<-mean(Y2) ; s<-sd(Y2)
> hist(Y2,breaks=0.1*(-45:45))
> lines(0.1*(-45:45),0.1*10*dnorm(0.1*(-45:45),m,s))
>
> Do it once, and the fit looks pretty good. However, if you
> repeat the above commands several times, you will observe
> that, near the peak of the curve, there is a definite
> tendency for the peak of the histogram to lie below the peak
> of the fitted curve. This illustrates the fact that you are
> looking at a superposition of two Normal curves, with identical
> (since you have p=0.5 in the Binomial sample) proportions
> and variances, but slightly shifted relative to each other
> (by an amount r which is in magnitude less than 1). This
> superposition is flatter at the top than any single Normal
> curve would be, which is being reflected in the "deficiency"
> of the histogram relative to the fitted curve
>
> You can also see this theoretically, since your Y2 is
>
> Y2 = A*X1 + B*Y
>
> so that
>
> P[Y2 <= y] = p*P[B*Y <= y] + (1-p)*P[B*Y <= y - A]
>
> = p*P[Y <= y/B] + (1-p)*P[Y <= (y-A)/B]
>
> where p is the Binomial P[X1 = 1].
>
> Hence the density function of Y2 is the derivative of this
> w.r.to y, namely
>
> p*f(y/B)/B + (1-p)*f((y-A)/B)/B
>
> where f(x) is the density function of the standard N(0,1).
>
> While in the case of your example (see histograms) the
> distribution of Y2 might be close enough to Normal for
> practical purposes (but this depends on your practical
> purpose), it is nonetheless better to try to avoid the
> theoretical difficulty if possible.
>
> So I'd suggest experimenting on the following lines.
>
> 1. Let X1 be a sample of size N using rbinom(N,1,p)
>(where, in general, p need not be 0.5)
>
> 2. Let Y be a sample of size N using rnorm(N,mu,sigma)
>(and again, in general, mu need not be 0 nor sigma 1).
>
> This is
[R] nls error in formula
Hi, I'm a new R user, with a lot of questions. At the moment I'm stoped on an error traying to fit a model: > x <- sandeel ## numeric data (2500-6) > y <- Noss ## numeric data (0-1.2) > A <- 0.8 > B <- 0.6 > C <- 1/4 > nls( y ~ A-B*exp(-C*x)) Error in match.call(definition, call, expand.dots) : .Primitive... is not a function I'm not sure if the error is due to the formula sytax or something else. I'll appreciate any help Sylvia _ Do You Yahoo!? La mejor conexión a internet y 25MB extra a tu correo por $100 al mes. http://net.yahoo.com.mx __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] generalized linear mixed models - how to compare?
Liset
Dat moet lukken en hoe sneller je het anlevert hoe eerder je resultaten terug
ziet.
Dus, Ik wacht af.
groet Peter S
Date sent: Sun, 17 Apr 2005 18:07:28 +0100 (BST)
From: Prof Brian Ripley <[EMAIL PROTECTED]>
To: Deepayan Sarkar <[EMAIL PROTECTED]>
Subject:Re: [R] generalized linear mixed models - how to
compare?
Copies to: r-help@stat.math.ethz.ch,
Nestor Fernandez <[EMAIL PROTECTED]>
> On Sun, 17 Apr 2005, Deepayan Sarkar wrote:
>
> > On Sunday 17 April 2005 08:39, Nestor Fernandez wrote:
>
> >> I want to evaluate several generalized linear mixed models, including
> >> the null model, and select the best approximating one. I have tried
> >> glmmPQL (MASS library) and GLMM (lme4) to fit the models. Both result
> >> in similar parameter estimates but fairly different likelihood
> >> estimates.
> >> My questions:
> >> 1- Is it correct to calculate AIC for comparing my models, given that
> >> they use quasi-likelihood estimates? If not, how can I compare them?
> >> 2- Why the large differences in likelihood estimates between the two
> >> procedures?
> >
> > The likelihood reported by glmmPQL is wrong, as it's the likelihood of
> > an incorrect model (namely, an lme model that approximates the correct
> > glmm model).
>
> Actually glmmPQL does not report a likelihood. It returns an object of
> class "lme", but you need to refer to the reference for how to interpret
> that. It *is* support software for a book.
>
> > GLMM uses (mostly) the same procedure to get parameter estimates, but as
> > a final step calculates the likelihood for the correct model for those
> > estimates (so the likelihood reported by it should be fairly reliable).
>
> Well, perhaps but I need more convincing. The likelihood involves many
> high-dimensional non-analytic integrations, so I do not see how GLMM can
> do those integrals -- it might approximate them, but that would not be
> `calculates the likelihood for the correct model'. It would be helpful to
> have a clarification of this claim. (Our experiments show that finding an
> accurate value of the log-likelihood is difficult and many available
> pieces of software differ in their values by large amounts.)
>
> Further, since neither procedure does ML fitting, this is not a maximized
> likelihood as required to calculate an AIC value. And even if it were,
> you need to be careful as often one GLMM is a boundary value for another,
> in which case the theory behind AIC needs adjustment.
>
> --
> Brian D. Ripley, [EMAIL PROTECTED]
> Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
> University of Oxford, Tel: +44 1865 272861 (self)
> 1 South Parks Road, +44 1865 272866 (PA)
> Oxford OX1 3TG, UKFax: +44 1865 272595
>
> __
> R-help@stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Stichting NIVEL / NETHERLANDS INSTITUTE OF PRIMARY HEALTH CARE
Peter Spreeuwenberg
P.O. Box 1568
3500 BN Utrecht
Netherland
E-mail : [EMAIL PROTECTED]
Direct: +31-30-2729678
General : +31-30-2729700
Fax : +31-30-2729729
Web-site http://www.nivel.nl
DISCLAIMER__
This message contains information that may be privileged or\...{{dropped}}
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] R2.0.1 for Mac OS X 10.3 problem
I have the same problem, in Windows, and I think the .Rdata is not corrupted. Load R without loading the.Rdata file - move or rename it. Then library(lme4) and load the .Rdata - it should work. cheers, Horacio -- Horacio Montenegro PhD Student in Genetics and Molecular Biology Universidade Estadual de Campinas (Unicamp) phone: +55 19 3788 1151 --- Prof Brian Ripley <[EMAIL PROTECTED]> wrote: > It did not `break R': just rename your .RData which > appears to have become > corrupted. > > Please always start R with --vanilla before > concluding it is broken. > > On Mon, 18 Apr 2005, Greg Arnold wrote: > > > This combination was operating satisfactorily > until I tried updating lme4 and > > Matrix. My attempts to do this ultimately broke > R. The R console appears > > briefly, then collapses. I have tried > downloading and reinstalling R > > without success. Typing 'R' into Terminal gives > the error message bring up > > the usual introduction, then the error message: > > Error in methods:::mlistMetaName(mi, ns) : > >The methods object name for "coerce" must > include the name of the > > package that contains the generic function, but > there is no generic function > > of this name > > Fatal error: unable to restore saved data in > .RData > > -- > Brian D. Ripley, > [EMAIL PROTECTED] > Professor of Applied Statistics, > http://www.stats.ox.ac.uk/~ripley/ > University of Oxford, Tel: +44 1865 > 272861 (self) > 1 South Parks Road, +44 1865 > 272866 (PA) > Oxford OX1 3TG, UKFax: +44 1865 > 272595 > > __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] the graph gallery strikes back
Yes, it works for me now. Thanks. Very impressive I might add. On 4/18/05, Romain Francois <[EMAIL PROTECTED]> wrote: > Oops :) > > That should be allright now. > > Romain > > Le 18.04.2005 13:52, Naji a écrit : > > >Hi Romain, > > > > > >Great job.. Thanks a lot. > >When pointing at the graph or the link (ie Kernel density estimator in R², > >Perspective plot and contour plot, etc) the browser is sending back a SQL > >error (Safari, MacOSX3.8) > > > >Erreur SQL! > >Unknown column 'qualification' in 'field list' > > > >Best regards > >Naji > > > > > >Le 18/04/05 11:29, « Romain Francois » <[EMAIL PROTECTED]> a écrit : > > > > > > > >>Hello useRs and helpRs, > >> > >>Some time ago, in a gallaxy far away (here is the thread : > >>http://finzi.psych.upenn.edu/R/Rhelp02a/archive/46532.html ) we > >>discussed about a graph gallery showing the power of R in that domain. I > >>did some work around that, and there is a (pretty advanced) draft here : > >> > >>http://addictedtor.free.fr/graphiques/displayGallery.php > >> > >>For instance, there are some of my graphs, some of Eric Lecoutre's and > >>some coming from demo(graphics), demo(image), demo(persp) and so on. > >> > >>Pretty soon, I'll add the possibility to : > >>- add graphs dynamically from the web > >>- give a mark to each graph (maybe we can do some stats :) ) > >>- link R function to their help pages from http://finzi.psych.upenn.edu/R > >>- add references to publications > >>- whatever > >> > >>Please take a look and tell me what should be improved. May the foRce be > >>with you. > >> > >>Romain. > >> > >>PS1 : Regarding graphs from demo(*), I wrote that author is R > >>Development Core Team, maybe those were done by a precise person in the > >>Core, in that case tell me and I'll change it. > >> > >>PS2 : Source code highlighting was done by the highlight software : ( > >>http://www.andre-simon.de/ ) suggested to me by Eric Lecoutre. > >> > >> > >> > > > > > > > > > > > > > > -- > ~ > ~~ Romain FRANCOIS - http://addictedtor.free.fr ~~ > Etudiant ISUP - CS3 - Industrie et Services > ~~http://www.isup.cicrp.jussieu.fr/ ~~ > Stagiaire INRIA Futurs - Equipe SELECT > ~~ http://www.inria.fr/recherche/equipes/select.fr.html~~ > ~ > > __ > R-help@stat.math.ethz.ch mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html > __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] lmer question
Markus Bean wrote:
Hi --
I'm using lmer for binomial data. I am trying to
replicate estimates provided by Agresti (2002,
Categorical data analysis, Wiley) using abortion data
in table 10.13 (estimates provided in table 12.3 p.
505).
I fit the same model using these three commands:
a1 <- lmer(resp ~ sex + option1 + option2 + (1|id),
data=abort,family=binomial, method = c("AGQ"))
a2 <- lmer(resp ~ sex + option1 + option2 + (1|id),
data=abort,family=binomial, method = c("Laplace"))
a3 <- lmer(resp ~ sex + option1 + option2 + (1|id),
data=abort,family=binomial, method = c("PQL"))
All three methods provide the exact same estimates
(which should not be the case), and the estimates are
incorrect. I know the data are correctly entered
because I obtain correct estimates with gllamm in
Stata.
I am I doing something wrong here in my commands, or
is the lmer module not implementing AGQ and Laplace
properly with this version?
The latter. Actually it is not implementing them at all in this
version. I will upload a new release of the lme4 package later this
week and insert appropriate messages.
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] the graph gallery strikes back
Hello, I'm not sure i got you right, i added a link to index on the top right corner of each page. Is that what you meant ? Romain Le 18.04.2005 14:33, BXC (Bendix Carstensen) a écrit : On my windows system, I cannot get back to the overview page once I clicked on one of the items. Is this yours or my problem? Bendix -- Bendix Carstensen Senior Statistician Steno Diabetes Center Niels Steensens Vej 2 DK-2820 Gentofte Denmark tel: +45 44 43 87 38 mob: +45 30 75 87 38 fax: +45 44 43 07 06 [EMAIL PROTECTED] www.biostat.ku.dk/~bxc -- -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Romain Francois Sent: Monday, April 18, 2005 11:29 AM To: RHELP; [EMAIL PROTECTED] Subject: [R] the graph gallery strikes back Hello useRs and helpRs, Some time ago, in a gallaxy far away (here is the thread : http://finzi.psych.upenn.edu/R/Rhelp02a/archive/46532.html ) we discussed about a graph gallery showing the power of R in that domain. I did some work around that, and there is a (pretty advanced) draft here : http://addictedtor.free.fr/graphiques/displayGallery.php For instance, there are some of my graphs, some of Eric Lecoutre's and some coming from demo(graphics), demo(image), demo(persp) and so on. Pretty soon, I'll add the possibility to : - add graphs dynamically from the web - give a mark to each graph (maybe we can do some stats :) ) - link R function to their help pages from http://finzi.psych.upenn.edu/R - add references to publications - whatever Please take a look and tell me what should be improved. May the foRce be with you. Romain. PS1 : Regarding graphs from demo(*), I wrote that author is R Development Core Team, maybe those were done by a precise person in the Core, in that case tell me and I'll change it. PS2 : Source code highlighting was done by the highlight software : ( http://www.andre-simon.de/ ) suggested to me by Eric Lecoutre. -- ~ ~~ Romain FRANCOIS - http://addictedtor.free.fr ~~ Etudiant ISUP - CS3 - Industrie et Services ~~http://www.isup.cicrp.jussieu.fr/ ~~ Stagiaire INRIA Futurs - Equipe SELECT ~~ http://www.inria.fr/recherche/equipes/select.fr.html~~ ~ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] the graph gallery strikes back
Great work! Nice collection! For me everything worked, although I did't click on all the graphs! Thans! - Original Message - From: "Romain Francois" <[EMAIL PROTECTED]> To: "RHELP" ; <[EMAIL PROTECTED]> Sent: Monday, April 18, 2005 11:29 AM Subject: [R] the graph gallery strikes back Hello useRs and helpRs, Some time ago, in a gallaxy far away (here is the thread : http://finzi.psych.upenn.edu/R/Rhelp02a/archive/46532.html ) we discussed about a graph gallery showing the power of R in that domain. I did some work around that, and there is a (pretty advanced) draft here : http://addictedtor.free.fr/graphiques/displayGallery.php For instance, there are some of my graphs, some of Eric Lecoutre's and some coming from demo(graphics), demo(image), demo(persp) and so on. Pretty soon, I'll add the possibility to : - add graphs dynamically from the web - give a mark to each graph (maybe we can do some stats :) ) - link R function to their help pages from http://finzi.psych.upenn.edu/R - add references to publications - whatever Please take a look and tell me what should be improved. May the foRce be with you. Romain. PS1 : Regarding graphs from demo(*), I wrote that author is R Development Core Team, maybe those were done by a precise person in the Core, in that case tell me and I'll change it. PS2 : Source code highlighting was done by the highlight software : ( http://www.andre-simon.de/ ) suggested to me by Eric Lecoutre. -- ~ ~~ Romain FRANCOIS - http://addictedtor.free.fr ~~ Etudiant ISUP - CS3 - Industrie et Services ~~http://www.isup.cicrp.jussieu.fr/ ~~ Stagiaire INRIA Futurs - Equipe SELECT ~~ http://www.inria.fr/recherche/equipes/select.fr.html~~ ~ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] R-2.1.0 is released
I've rolled up R-2.1.0.tar.gz a short while ago. This version contains
several substantial changes and additions, perhaps most notably the
UTF8 and Internationalization changes. See the full list of changes
below.
You can get it from
http://cran.r-project.org/src/base/R-2/R-2.1.0.tar.gz
or wait for it to be mirrored at a CRAN site nearer to you. Binaries
for various platforms will appear in due course.
There is also a version split for floppies (does anyone use these any
more?).
For the R Core Team
Peter Dalgaard
These are the md5sums for the freshly created files, in case you wish
to check that they are uncorrupted:
94d55d512a9ba36caa9b7df079bae19f COPYING
d8045f3b8f929c1cb29a1e3fd737b499 COPYING.LIB
70447ae7f2c35233d3065b004aa4f331 INSTALL
8b78e12b100a6834fdada2dfaab98ab0 NEWS
88bbd6781faedc788a1cbd434194480c ONEWS
4f004de59e24a52d0f500063b4603bcb OONEWS
e7f92ef0c2bc5b772a83f42a5ed504c9 R-2.1.0.tar.gz
d45b1084f91fcfeebbb69cc2dd9ed6e1 R-2.1.0.tar.gz-split.aa
18b34d82302a68a86510da6e0dd9a193 R-2.1.0.tar.gz-split.ab
d242b365ed1398031ff26d31a6da9c1c R-2.1.0.tar.gz-split.ac
b992e7d10baf662ede52a0c156f51d86 R-2.1.0.tar.gz-split.ad
a2c559b4b1c87f5815bec9e8e2b7bd31 R-2.1.0.tar.gz-split.ae
033ae9f596d74f9780f98c9b6cab31df R-2.1.0.tar.gz-split.af
ed0fd6939e8e84b0a4a2a01d92cc7f96 R-2.1.0.tar.gz-split.ag
2d1379175456b308328d9af09e2726a3 R-2.1.0.tar.gz-split.ah
bbb8fd837e27c931b9921548ee4c5c28 R-2.1.0.tar.gz-split.ai
e7f92ef0c2bc5b772a83f42a5ed504c9 R-latest.tar.gz
56a780cdec835c5c598f8dfc0738f7f3 README
Here is the relevant bit of the NEWS file:
CHANGES IN R VERSION 2.1.0
USER-VISIBLE CHANGES
o box plots {by boxplot() or bxp()} now have the median line three
times the normal line width in order to distinguish it from the
quartile ones.
o Unix-alike versions of R can now be used in UTF-8 locales on
suitably equipped OSes. See the internationalization section
below.
o The meaning of 'encoding' for a connection has changed: See the
internationalization section below.
o There has been some rationalization of the format of
warning/error messages, to make them easier to translate.
Generally names of functions and arguments are single-quoted,
and classes double-quoted.
o Reading text files with embedded "\" (as in Windows file names)
may now need to use scan(* , allowEscapes = FALSE), see also below.
NEW FEATURES
o %% now warns if its accuracy is likely to be affected by lack
of precision (as in 1e18 %% 11, the unrealistic expectation of
PR#7409), and tries harder to return a value in range when it is.
o abbreviate() now warns if used with non-ASCII chars, as the
algorithm is designed for English words.
o The default methods for add1() and drop1() check for changes
in the number of cases in use.
The "lm" and "glm" methods for add1() quoted the model
on the original fitted values when using (with a warning) a
smaller set of cases for the expanded models.
o Added alarm() function to generate a bell or beep or visual alert.
o all/any() now attempt to coerce their arguments to logical, as
documented in the Blue Book. This means e.g. any(list()) works.
o New functions for multivariate linear models: anova.mlm(),
SSD(), estVar(), mauchley.test() (for sphericity).
vcov() now does something more sensible for "mlm" class objects.
o as.data.frame.table() has a new argument 'responseName'
(contributed by Bill Venables).
o as.dist() and cophenetic() are now generic, and the latter has a
new method for objects of class "dendrogram".
o as.ts() is now generic.
o binomial() has a new "cauchit" link (suggested by Roger Koenker).
o chisq.test() has a new argument 'rescale.p'. It is now possible
to simulate (slowly) the P value also in the 1D case
(contributed by Rolf Turner).
o choose(n,k) and lchoose(.) now also work for arbitrary (real) n
in accordance with the general binomial theorem. choose(*,k)
is more accurate (and faster) for small k.
o Added colorRamp() and colorRampPalette() functions for color
interpolation.
o colSums()/rowSums() now allow arrays with a zero-length extent
(requested by PR#7775).
o confint() has stub methods for classes "glm" and "nls" that
invoke those in package MASS. This avoids using the "lm"
method for "glm" objects if MASS is not attached.
confint() has a default method using asymptotic normality.
o contr.SAS() has been moved from the 'nlme' package to the
'stats' package.
o New function convertColors() maps between color spaces.
colorRamp() uses it.
o The cov() function in the non-Pearson cases now ranks data after
Re: [R] the graph gallery strikes back
Hello Hans, For the moment, you can't put yourself the graph in it. I'll do it for you. Just send me the source code and i'll put your graph when i found the time (probably tonight). You may also want to add comments on Andrew curves. Romain Le 18.04.2005 13:37, Hans Werner Borchers a écrit : Dear Francois Romain, thanks for providing the R Graphics Gallery. Actually, I would like to provide a graph of "Andrews curves", probably for the Iris data set. Also, one could add some of the beautiful 3D pictures generated with the RGL package. Unfortunately, it appears to be impossible to retrieve the code behind: "Erreur SQL! Unknown column 'qualification' in 'field list'" Best regards, Hans Werner Borchers. -- ~ ~~ Romain FRANCOIS - http://addictedtor.free.fr ~~ Etudiant ISUP - CS3 - Industrie et Services ~~http://www.isup.cicrp.jussieu.fr/ ~~ Stagiaire INRIA Futurs - Equipe SELECT ~~ http://www.inria.fr/recherche/equipes/select.fr.html~~ ~ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] the graph gallery strikes back
Oops :) That should be allright now. Romain Le 18.04.2005 13:52, Naji a écrit : Hi Romain, Great job.. Thanks a lot. When pointing at the graph or the link (ie Kernel density estimator in R², Perspective plot and contour plot, etc) the browser is sending back a SQL error (Safari, MacOSX3.8) Erreur SQL! Unknown column 'qualification' in 'field list' Best regards Naji Le 18/04/05 11:29, « Romain Francois » <[EMAIL PROTECTED]> a écrit : Hello useRs and helpRs, Some time ago, in a gallaxy far away (here is the thread : http://finzi.psych.upenn.edu/R/Rhelp02a/archive/46532.html ) we discussed about a graph gallery showing the power of R in that domain. I did some work around that, and there is a (pretty advanced) draft here : http://addictedtor.free.fr/graphiques/displayGallery.php For instance, there are some of my graphs, some of Eric Lecoutre's and some coming from demo(graphics), demo(image), demo(persp) and so on. Pretty soon, I'll add the possibility to : - add graphs dynamically from the web - give a mark to each graph (maybe we can do some stats :) ) - link R function to their help pages from http://finzi.psych.upenn.edu/R - add references to publications - whatever Please take a look and tell me what should be improved. May the foRce be with you. Romain. PS1 : Regarding graphs from demo(*), I wrote that author is R Development Core Team, maybe those were done by a precise person in the Core, in that case tell me and I'll change it. PS2 : Source code highlighting was done by the highlight software : ( http://www.andre-simon.de/ ) suggested to me by Eric Lecoutre. -- ~ ~~ Romain FRANCOIS - http://addictedtor.free.fr ~~ Etudiant ISUP - CS3 - Industrie et Services ~~http://www.isup.cicrp.jussieu.fr/ ~~ Stagiaire INRIA Futurs - Equipe SELECT ~~ http://www.inria.fr/recherche/equipes/select.fr.html~~ ~ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Rcmdr's problem in East Asian Languages Version
The issue here is not the `East Asian Languages Version' but the locale you are using, which you have not told us. Try using the C locale or an English locale: Rcmdr works there for me. Also be sure to tell us the version of the package (here Rcmdr) you are using. The posting guide does ask you to report package-specific problems to the maintainer, and I have sent a more detailed reply to him copied to you. On Mon, 18 Apr 2005, ronggui wrote: maybe this is a problem specific to the East Asian Languages Version.when i reinstall R without checking the "Version for East Asian Languages" option ,it works.but if i install R with the "Version for East Asian Languages" option, it show the following error msg. library(Rcmdr) Loading required package: tcltk Loading Tcl/Tk interface ... done Error in parse(file, n, text, prompt) : syntax error on line 4857 Error: unable to load R code in package 'Rcmdr' Error: package/namespace load failed for 'Rcmdr' [ 2.1.0 beta ] -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] the graph gallery strikes back
On 4/18/05, Romain Francois <[EMAIL PROTECTED]> wrote: > Hello useRs and helpRs, > > Some time ago, in a gallaxy far away (here is the thread : > http://finzi.psych.upenn.edu/R/Rhelp02a/archive/46532.html ) we > discussed about a graph gallery showing the power of R in that domain. I > did some work around that, and there is a (pretty advanced) draft here : > > http://addictedtor.free.fr/graphiques/displayGallery.php > > For instance, there are some of my graphs, some of Eric Lecoutre's and > some coming from demo(graphics), demo(image), demo(persp) and so on. > > Pretty soon, I'll add the possibility to : > - add graphs dynamically from the web > - give a mark to each graph (maybe we can do some stats :) ) > - link R function to their help pages from http://finzi.psych.upenn.edu/R > - add references to publications > - whatever > > Please take a look and tell me what should be improved. May the foRce be > with you. > > Romain. > > PS1 : Regarding graphs from demo(*), I wrote that author is R > Development Core Team, maybe those were done by a precise person in the > Core, in that case tell me and I'll change it. > > PS2 : Source code highlighting was done by the highlight software : ( > http://www.andre-simon.de/ ) suggested to me by Eric Lecoutre. > When I click on any of the graphs or the title of a graph I get this: Erreur SQL! Unknown column 'qualification' in 'field list' I am using Windows XP and Internet Explorer 6.0 (with service pack 2). __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] R2.0.1 for Mac OS X 10.3 problem
It did not `break R': just rename your .RData which appears to have become corrupted. Please always start R with --vanilla before concluding it is broken. On Mon, 18 Apr 2005, Greg Arnold wrote: This combination was operating satisfactorily until I tried updating lme4 and Matrix. My attempts to do this ultimately broke R. The R console appears briefly, then collapses. I have tried downloading and reinstalling R without success. Typing 'R' into Terminal gives the error message bring up the usual introduction, then the error message: Error in methods:::mlistMetaName(mi, ns) : The methods object name for "coerce" must include the name of the package that contains the generic function, but there is no generic function of this name Fatal error: unable to restore saved data in .RData -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] how to use sink.number()
On Mon, 18 Apr 2005, Michael S wrote:
Dear All R-helper,
my question is how to use sink.number.
if I want to label the output was sinked by sink function how should I do ?
example:
zz <- file("c:\\sinktest1.txt",open="wt")
sink(zz)
g<- 1:10
cat(g,"\n")
f<-list(1:100)
h<-capture.output(f)
cat(h)
sink()
I want there are marks in "sinktest1.txt" files, let me know the number of
output I get(in this example, there should be 2 mark or number ,becuase I
using two cat function)
That is nothing to do with sink.number():
'sink.number()' reports how many diversions are in use.
'sink.number(type = "message")' reports the number of the
connection currently being used for error messages.
If you don't understand that, please consult the references on the help
page.
--
Brian D. Ripley, [EMAIL PROTECTED]
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UKFax: +44 1865 272595
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Rcmdr's problem in East Asian Languages Version
maybe this is a problem specific to the East Asian Languages Version.when i reinstall R without checking the "Version for East Asian Languages" option ,it works.but if i install R with the "Version for East Asian Languages" option, it show the following error msg. > library(Rcmdr) Loading required package: tcltk Loading Tcl/Tk interface ... done Error in parse(file, n, text, prompt) : syntax error on line 4857 Error: unable to load R code in package 'Rcmdr' Error: package/namespace load failed for 'Rcmdr' > version _ platform i386-pc-mingw32 arch i386 os mingw32 system i386, mingw32 status beta major2 minor1.0 year 2005 month04 day 16 language R __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Invalid *.Rd in package {was "Rd.sty problems"}
> "Federico" == Federico Calboli <[EMAIL PROTECTED]>
> on Mon, 18 Apr 2005 10:33:22 +0100 writes:
Federico> I am trying to build a new R package to submit,
Federico> but it's failing to create a tex manual:
Federico> R CMD check Biodem
Federico> * checking for working latex ... OK
Federico> * using log directory
Federico> '/home/greatsage/Fede/R-packages/temp/Biodem.Rcheck'
Federico> * checking for file 'Biodem/DESCRIPTION' ... OK
Federico> * checking if this is a source package ... OK
Federico> * Installing *source* package 'Biodem' ...
Federico> ** R
Federico> ** data
Federico> ** help
Building/Updating help pages for package 'Biodem'
Federico> Formats: text html latex example
Federico> * DONE (Biodem)
Federico> * checking package directory ... OK
. (everything OK)
Federico> * creating Biodem-manual.tex ... OK
Federico> * checking Biodem-manual.tex ... ERROR
Federico> LaTeX errors when creating DVI version.
Federico> This typically indicates Rd problems.
(and I think I have never seen another situation: It will be a
problem in of your Biodem/man/*.Rd files !)
Federico> I cheked the log file in the .Rcheck dir, but it's
Federico> incomprehensible to me, so I simply copied and run
Federico> the tex file itself. The manual won't compile
Federico> lamenting it cannot find the Rd.sty file.
Federico> locate Rd.sty gives:
Federico> /usr/lib/R/share/texmf/Rd.sty
Federico> I am completely lost... any advice? do I have to
Federico> change some path? I am using R for Debian from the
Federico> latest .deb files for Sarge on x86 arch.
When 'R CMD check' latexs your file it adds something like
/usr/lib/R/share/texmf/Rd.sty to the TEXINPUTS (path) used
and you could do so as well.
Alternatively, use
R CMD Rd2dvi
directly.
If this doesn't help you, you may need to look at the latex
*.log file and find someone to whom it's not incomprehensible,
possibly by putting it up on the web asking for help on R-devel
{because the whole issue will have become too technical for R-help}.
Martin
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] R2.0.1 for Mac OS X 10.3 problem
This combination was operating satisfactorily until I tried updating lme4 and Matrix. My attempts to do this ultimately broke R. The R console appears briefly, then collapses. I have tried downloading and reinstalling R without success. Typing 'R' into Terminal gives the error message bring up the usual introduction, then the error message: Error in methods:::mlistMetaName(mi, ns) : The methods object name for "coerce" must include the name of the package that contains the generic function, but there is no generic function of this name Fatal error: unable to restore saved data in .RData Can anyone help? Greg __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Rd.sty problems.
Hi All, I am trying to build a new R package to submit, but it's failing to create a tex manual: R CMD check Biodem * checking for working latex ... OK * using log directory '/home/greatsage/Fede/R-packages/temp/Biodem.Rcheck' * checking for file 'Biodem/DESCRIPTION' ... OK * checking if this is a source package ... OK * Installing *source* package 'Biodem' ... ** R ** data ** help >>> Building/Updating help pages for package 'Biodem' Formats: text html latex example Fst texthtmllatex example Laskertexthtmllatex example Mal.eqtexthtmllatex example N texthtmllatex example P texthtmllatex example Phi texthtmllatex example S texthtmllatex example col.sto texthtmllatex example hedrick texthtmllatex example mal.cond texthtmllatex example mar.iso texthtmllatex example mtx.exp texthtmllatex example r.pairs texthtmllatex example raw.mig texthtmllatex example regional.rnd.iso texthtmllatex example rel.cond texthtmllatex example rel.phi texthtmllatex example rri texthtmllatex example sur.freq texthtmllatex example sur.inbr texthtmllatex example surnames texthtmllatex example sym.P texthtmllatex example uri texthtmllatex example valleytexthtmllatex example * DONE (Biodem) * checking package directory ... OK * checking for portable file names ... OK * checking for sufficient/correct file permissions ... OK * checking package dependencies ... OK * checking index information ... OK * checking package subdirectories ... OK * checking R files for syntax errors ... OK * checking R files for library.dynam ... OK * checking S3 generic/method consistency ... OK * checking replacement functions ... OK * checking foreign function calls ... OK * checking Rd files ... OK * checking for missing documentation entries ... OK * checking for code/documentation mismatches ... OK * checking Rd \usage sections ... OK * creating Biodem-Ex.R ... OK * checking examples ... OK * creating Biodem-manual.tex ... OK * checking Biodem-manual.tex ... ERROR LaTeX errors when creating DVI version. This typically indicates Rd problems. I cheked the log file in the .Rcheck dir, but it's incomprehensible to me, so I simply copied and run the tex file itself. The manual won't compile lamenting it cannot find the Rd.sty file. locate Rd.sty gives: /usr/lib/R/share/texmf/Rd.sty I am completely lost... any advice? do I have to change some path? I am using R for Debian from the latest .deb files for Sarge on x86 arch. Regards, Federico Calboli -- Federico C. F. Calboli Department of Epidemiology and Public Health Imperial College, St Mary's Campus Norfolk Place, London W2 1PG Tel +44 (0)20 7594 1602 Fax (+44) 020 7594 3193 f.calboli [.a.t] imperial.ac.uk f.calboli [.a.t] gmail.com __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] the graph gallery strikes back
Hello useRs and helpRs, Some time ago, in a gallaxy far away (here is the thread : http://finzi.psych.upenn.edu/R/Rhelp02a/archive/46532.html ) we discussed about a graph gallery showing the power of R in that domain. I did some work around that, and there is a (pretty advanced) draft here : http://addictedtor.free.fr/graphiques/displayGallery.php For instance, there are some of my graphs, some of Eric Lecoutre's and some coming from demo(graphics), demo(image), demo(persp) and so on. Pretty soon, I'll add the possibility to : - add graphs dynamically from the web - give a mark to each graph (maybe we can do some stats :) ) - link R function to their help pages from http://finzi.psych.upenn.edu/R - add references to publications - whatever Please take a look and tell me what should be improved. May the foRce be with you. Romain. PS1 : Regarding graphs from demo(*), I wrote that author is R Development Core Team, maybe those were done by a precise person in the Core, in that case tell me and I'll change it. PS2 : Source code highlighting was done by the highlight software : ( http://www.andre-simon.de/ ) suggested to me by Eric Lecoutre. -- ~ ~~ Romain FRANCOIS - http://addictedtor.free.fr ~~ Etudiant ISUP - CS3 - Industrie et Services ~~http://www.isup.cicrp.jussieu.fr/ ~~ Stagiaire INRIA Futurs - Equipe SELECT ~~ http://www.inria.fr/recherche/equipes/select.fr.html~~ ~ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] [R-pkgs] RWinEdt 1.7-0 released
Dear useRs,
RWinEdt_1.7-0 has been released.
Windows users of R-2.1.0 (which is scheduled to be released today) who
are going to use a translated (i.e. non-english) version of RGui
together with WinEdt will have to upgrade.
It is not advisable to upgrade RWinEdt for R versions < 2.1.0.
RWinEdt can be used in form of an R package (e.g. by saying
install.packages("RWinEdt") within R-2.1.0) and as a manually
installable WinEdt Plug-In.
Uwe Ligges
___
R-packages mailing list
[EMAIL PROTECTED]
https://stat.ethz.ch/mailman/listinfo/r-packages
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Mac GUI Slow?
Hi, I have R version 2.0.1 on my PowerBook G4 1.5Ghz, 512RAM, running OS X 10.3.9. I have noticed that the "official" R GUI runs considerably slower than R in the terminal and also JGR. Whilst the differences in speed are unimportant for simple parametric stats, it becomes a bit of a pig when doing non-parametric stats e.g. neural nets, feature selection algorithms. In all my experiments I have set my CPU speed to "maximum" in system preferences. I have also looked at the CPU usage for each implementation: on average R in the terminal and JGR manage to get 98% of CPU usage, compared to only around 75% for the R GUI. I guess my questions are: 1) Why the difference in speed when its on the same machine? 2) Is there any way of tweaking R GUI to make it run faster, a la R in the terminal? Any help would be appreciated. Best regards, Sam. [[alternative text/enriched version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] how to use sink.number()
Dear All R-helper,
my question is how to use sink.number.
if I want to label the output was sinked by sink function how should I do ?
example:
zz <- file("c:\\sinktest1.txt",open="wt")
sink(zz)
g<- 1:10
cat(g,"\n")
f<-list(1:100)
h<-capture.output(f)
cat(h)
sink()
I want there are marks in "sinktest1.txt" files, let me know the number of
output I get(in this example, there should be 2 mark or number ,becuase I
using two cat function)
thanks
Michael
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
