Re: [R] Formatting Sweave in R-News
Arjun Ravi Narayan contact at arjunnarayan.com writes: I am editing a document for submission to the R-news newsletter, and in my article my Sweave code inserts a dynamically generated PDF report that my R program generates. Slightly off Arjuns topic is a problem with Schunk, Sinput and Soutput environments. They just use to much inline space. I have tweade this problem with sed (see bellow for Makefile content), but wonder if there is a better solution. Thanks, Gregor default: Sweave fixTex texi2pdf --clean wrapper.tex evince wrapper.pdf Sweave: # Sweave myPaper.Rnw R CMD Sweave myPaper.Rnw fixTex: Sweave # Change all S* environments to smallverbatim @cat myPaper.tex | sed -e '/\\begin{Sinput}/d' \ -e '/\\end{Sinput}/d' \ -e '/\\begin{Soutput}/d' \ -e '/\\end{Soutput}/d' \ -e 's/\\begin{Schunk}/\\begin{smallverbatim}/g' \ -e 's/\\end{Schunk}/\\end{smallverbatim}/g' \ myPaper2.tex @mv -f myPaper2.tex myPaper.tex __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Formatting Sweave in R-News
Hi Duncan, Duncan Murdoch wrote: Slightly off Arjuns topic is a problem with Schunk, Sinput and Soutput environments. They just use to much inline space. I have tweade this problem with sed (see bellow for Makefile content), but wonder if there is a better solution. You can change the spacing by redefining those environments. I use this: % This removes the extra spacing after code and output chunks in Sweave, % but keeps the spacing around the whole block. \fvset{listparameters={\setlength{\topsep}{0pt}}} \renewenvironment{Schunk}{\vspace{\topsep}}{\vspace{\topsep}} Nice trick. Thank you. I am CCing to a friend that had the same issue. -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Facultywww: http://www.bfro.uni-lj.si/MR/ggorjan Zootechnical Department blog: http://ggorjan.blogspot.com Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale fax: +386 (0)1 72 17 888 Slovenia, Europetel: +386 (0)1 72 17 861 __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] rv package, rvnorm function
John P. Burkett burkett at uri.edu writes: In an attempt to learn to use the rv package, I have been working through the examples in Jouni Kerman and Andrew Gelman's Using Random Variables to Manipulate and Summarize Simulations in R (July 4, 2007). Perhaps Jouni can help you? __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] residual plots for lmer in lme4 package
John Maindonald john.maindonald at anu.edu.au writes: ... The issue of checking for normality of effects in multi-level models has not been very much researched, as far as I can tell. The function residuals() gives residuals that adjust for all except the highest level of random effects. Depending on the relative magnitudes of the variance components, whether or not these residuals are anywhere near normal may not be of much or any consequence.ž For what it is worth I have came across this paper just recently: http://www3.interscience.wiley.com/cgi-bin/abstract/114280441 Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] R2WinBUGS results not different with different runs
toby909 at gmail.com writes: Gregor Gorjanc wrote: toby909 at gmail.com writes: Is this a specialty with R2WinBUGS? Does it have something to do with the seed value? Isnt the seed value reset everytime I restart winbugs? I always have the same seed if I start WinBUGS multiple times. So you get exactly the same chain, numerically, when rerunning the same model, with the same number of iterations, everything the same.? I just tried now with Gelman's schools example from bugs() help page. I runned the same job twice, with exactly the same initial values. Notes that WinBUGS uses always the same starting seed. I got this schools.sim$last.values [[1]] [[1]]$theta [1] 23.700 10.060 12.760 13.090 1.693 14.390 7.599 3.961 [[1]]$mu.theta [1] 25.03 [[1]]$sigma.theta [1] 20.08 schools.sim2$last.values [[1]] [[1]]$theta [1] 23.700 10.060 12.760 13.090 1.693 14.390 7.599 3.961 [[1]]$mu.theta [1] 25.03 [[1]]$sigma.theta [1] 20.08 Wouldnt that be problematic if every researcher in the world who uses winbugs uses the same sequence of random numbers? R's random numbers are different each time, because the seed is linked to the clock in your PC. You can reset the seed if you want. Having the same seed is also nice for repeatability. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] R2WinBUGS results not different with different runs
toby909 at gmail.com writes: Is this a specialty with R2WinBUGS? Does it have something to do with the seed value? Isnt the seed value reset everytime I restart winbugs? I always have the same seed if I start WinBUGS multiple times. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] error in using R2WinBUGS on Ubuntu 6.10 Linux
J. Patrick Meyer meyerjp at jmu.edu writes: I'm using Wine 0.9.41, and the patched version of WinBUGS. Everything seems to run correctly. I get the right output from WinBUGS. I'm not sure to what the error message refers. I also get this messages, but that is comming from wine and can be safely ignored. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] icc from GLMM?
Shinichi Nakagawa S.Nakagawa at sheffield.ac.uk writes: ... I am a little confused which one to trust and use. Or there are no easy form to do this? I am guessing formula would change depending on what distribution you use and what link function as well? I want to calculate icc from GLMM with Poisson with log link function and also binomial with logit function. Could anybody help me please? Yes, you are right that ICC depends on assumed data distribution. While ICC is very handy in linear models it is not the case in GLMM. I suggest you take a look at the references bellow. There is also some online material by the same authors. Additionally, I remember that there were lively discussions about ICC on multilevel list at http://www.jiscmail.ac.uk/lists/multilevel.html Best wishes, Gregor @Article{Goldstein:2002, author = {Goldstein, H. and Browne, W. and Rabash, J.}, title ={Partitioning variation in multilevel models}, journal = {Understanding Statistics}, year = {2002}, volume = {1}, number = {4}, pages ={223--231}, keywords = {variance ratio, variance partition coefficient, intra-unit correlation, intra-class correlation, normal models, discrete models, random coefficient models} } @Article{Browne:2005, author = {Browne, W. J. and Subramanian, S. V. and Jones, K. and Goldstein, H.}, title ={Variance partitioning in multilevel logistic models that exhibit overdispersion}, journal = {J. R. Stat. Soc. A Stat. Soc.}, year = {2005}, volume = {168}, number = {3}, pages ={599--613}, doi = {10./j.1467-985X.2004.00365.x}, checked = {[2006-04-16]}, keywords = {heritability, ratios, intra-class correlation, intra-unit correlation, simulation, linearization, latent variable approach}, } __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Refactor all factors in a data frame
Hilmar Berger hilmar.berger at imise.uni-leipzig.de writes: ... So, is there any way in drop the unused factor levels from *all* factors of a data frame without import/export ? There is a generic drop.levels in gdata. Here is part of its help page: \code{drop.levels} is a generic function, where default method does nothing, while method for factor \code{s} drops all unused levels. There are also convenient methods for \code{list} and \code{data.frame}, where all unused levels are dropped in all factors (one by one) in a \code{list} or a \code{data.frame}. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] .C HoltWinters
sj ssj1364 at gmail.com writes: Hello, I would like to look at the compiled C code behind HoltWinters from the stats package. Is that possible? If so where do I find it? Get the R source tarball from CRAN unpack it and search for the C code. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] write fixed format
YIHSU CHEN yihsu.chen at ucmerced.edu writes: Dear R users; Is there a function in R that I can put text with proper alignments in a fixed format. For instance, if I have three fields: A, B and C, where both A and C are text with 3 characters and left alignment; B is a numeric one with 2 decimals and 3 integer space digits. How can I put the following row in a file? (note there is a space between a and 2, and after b.) (A=aaa, B=23.11 and C=bb) aaa 23.11bb Check write.fwf() in gdata package. It might help you, but you will have to create new column to have B and C together i.e. without any space between. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] setting a number of values to NA over a data.frame.
Jim Lemon jim at bitwrit.com.au writes: Hi John, You might have a look at toNA in the prettyR package. Wait for version 1.0-4, just uploaded, as I have fixed a bug in that function. There is also a set of generic functions exactly for such cases: unknownToNA(), NAToUnknown() and isUnknown() in gdata package. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] SQL statements (directly) in R
Wensui Liu liuwensui at gmail.com writes: chris, if you could create a package similar to proc sql in SAs, that will be so sweet. Nice task, but I would not suggest to go this way. I found it quite easy to work with R subsetting system after you get some practice and that is more than enough - at least for me. Maybe you should look into sqllite stuff that BioC guys are working on. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] lmer and estimation of p-values: error with mcmcpvalue()
Christoph Scherber Christoph.Scherber at agr.uni-goettingen.de writes: Dear Henric, Thanks, now it works; but how reliable are these estimates? Especially with p-values close to 0.05 it is of course important that the range of the estimates is not too large. I´ve just run several simulations, each of which yielding sometimes quite different p-values. I have not been using mcmcsamp(), but try increasing number of simulations. Additionally, since you have only single parameters, you can directly look at their posterior distributions and compute the areas under the curve. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] generate Binomial (not Binary) data
Marc Bernard bernarduse1 at yahoo.fr writes: I am looking for an R function or any other reference to generate a series of correlated Binomial (not a Bernoulli) data. The bindata library can do this for the binary not the binomial case. But binomial is just a sum of Bernoulli vars so this should be doable. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] ECB/Sidebar/R (Emacs) was: Re: kate editor for R
Ramon Diaz-Uriarte rdiaz02 at gmail.com writes: I had looked at ECB for C++ programming. It simply hadn't occurred to me that it would plug into ESS. I wasn't aware of it either until I attended Tony Rossini's tutorial at useR! 2006. Are there any materials for those who missed the course? Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] zero random effect sizes with binomial lmer
Daniel Ezra Johnson johnson4 at babel.ling.upenn.edu writes: 1) Yes, I have tweaked the data to show as clearly as I can that this is a bug, that a tiny change in initial conditions causes the collapse of a reasonable 'parameter' estimate. I would not call this a bug, since this is related to data and not to the software. I might be wrong! 2) mcmcsamp() does not work (currently) for binomial fitted models. Sorry, for wrong pointer. You could try with some other packages if they have support for binomial models with random effects. I would just try in BUGS -- take a look at R2WinBUGS or Brugs. 3) This is an issue of what happens when the sample is too small. For all larger data sets I have gotten a ranef variance between 0.05 and 1.00 or so. It makes no sense to say that as the data set gets smaller, the systematic variation between Items goes away. It doesn't, as I've shown. In the data I believe that when data gets smaller such parameters are harder to estimate and you can easily get 0 as MLE. above, certain Items were still 10+ times as likely (log-odds wise) to have Response==1 as others. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] zero random effect sizes with binomial lmer
Daniel Ezra Johnson johnson4 at babel.ling.upenn.edu writes: ... If one compares the random effect estimates, in fact, one sees that they are in the correct proportion, with the expected signs. They are just approximately eight orders of magnitude too small. Is this a bug? ... BLUPs are essentially shrinkage estimates, where shrinkage is determined with magnitude of variance. Lower variance more shrinkage towards the mean - zero in this case. So this is not a bug. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] zero random effect sizes with binomial lmer
Daniel Ezra Johnson johnson4 at babel.ling.upenn.edu writes: ... More broadly, is it hopeless to analyze this data in this manner, or else, what should I try doing differently? It would be very useful to be able to have reliable estimates of random effect sizes, even when they are rather small. ... You might try with mcmcsamp to get a better view of posterior distributions of your parameters. It might be the case that MLE for item variance is near 0, while its posterior distribution covers also values that are higher than 0. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] zero random effect sizes with binomial lmer
Daniel Ezra Johnson johnson4 at babel.ling.upenn.edu writes: If one compares the random effect estimates, in fact, one sees that they are in the correct proportion, with the expected signs. They are just approximately eight orders of magnitude too small. Is this a bug? BLUPs are essentially shrinkage estimates, where shrinkage is determined with magnitude of variance. Lower variance more shrinkage towards the mean - zero in this case. So this is not a bug. Gregor I doubled each data set by duplicating each Subject. There are now 46 subjects in each group instead of 23. So now A and B differ in 2 observations out of 322. The lmer resuls are sensible. Why would you want to do that? Doubling your data set is not what you want as you are hacking the analysis. You should definitely avoid this for real analysis! I know BLUPs are not like regular parameters, but doubling (or cutting in half) the data should not, in my opinion, cause this behavior. There is a lot of room for the original A variance estimate to be lower than B, maybe it should be 0.05, 0.01, 0.05, or whatever, but not .05. This is numerical procedure and if log-likelihood is flat then it might happen that algorithms give you such estimates. When you doubled the data log-likelihood gets more peaked shape and then it seems reasonable that estimates are more consistent. I understand that the between-Item variance is low, and probably it is no greater than what you would expect to occur by chance, but isn't that what hypothesis testing is for (anova, etc.)? Yes, but anova etc. is not a super tool. If you get parameter estimates that are essentially 0, do you still need hypothesis testing? Is my best way around the algorithm returning zero to do what I have done above, with my real data? That is, duplicate (or triplicate) Subjects to increase my data size, and thereby get a reasonably comparable (if wrong) estimate of the Item variance? Zero is not a reasonable estimate in any of these data sets. No! Try with mcmcsamp and then you might get better view of posterior distribution of item variance. I think that MLE is near zero with some support for positive values. Try to play with lmer i.e. with starting values etc. Additionally, why do you want to estimate variances separately for dataset A and B. Can you really suggest that they should be different between datasets? Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Genotypes are not all the same
Farrel Buchinsky fjbuch at gmail.com writes: Clearly g1 and g2 are not exactly the same even though they are the same class. I can make g2 behave as g1 if i type g3-genotype(g2) g3 then behaves as it should. The issue may have arrisen since g2 is just a subset of a much bigger object that was created using reshape command in the base package or maybe melt and cast in the reshape package. Can anyone enlighten me? genotype class behaves internally much like factor class. I guess that levels/alleles were dropped somewhere. You have to figure out where this happened. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] [Fwd: [AGDG-LIST:428] Summer Course in Guelph]
Original Message Subject: [AGDG-LIST:428] Summer Course in Guelph Date: Fri, 22 Dec 2006 09:12:24 -0500 From: Larry Schaeffer [EMAIL PROTECTED] Reply-To: [EMAIL PROTECTED] To: Animal Geneticist's Discussion [EMAIL PROTECTED] The Centre for Genetic Improvement of Livestock at the University of Guelph is pleased to announce a one week summer course entitled Animal Breeding Applications Using R to be given by Bill Szkotnicki and Larry Schaeffer from June 11 to June 15, 2007. Details will be available in the new year on the CGIL website: http://cgil.uoguelph.ca/ The content of the course will be: Introduction to R Getting data in and out Introductory statistical models Packages and graphics functions Distributions and simulations Developing your own functions Accessing the operating system The R API Database connectivity Programming in R using Fortran and C subroutines Animal breeding applications - Inbreeding routines - Animal models - Iteration on data - Maternal genetic effects - Multiple traits - Random regression - Threshold models - Genomic applications On the application form you will be able to indicate particular problems that you would like to see covered, and we will try to work them into the course. -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] change factor level 1,2,3 to red,blue,dark
Aimin Yan aiminy at iastate.edu writes: I am new to R. Maybe this is very simple question. I have a dataframe, there is column that is factor. This factor has three level that 1,2,3. Now I want to change these level(1,2,3) to level(red,blue,dark). Does anybody how to do this job? levels() help page has entry about this. Try x - factor(round(runif(n=10, min=1, max=3))) x [1] 3 1 1 3 1 2 2 2 1 2 Levels: 1 2 3 levels(x) - list(red=c(1), blue=c(2), dark=c(3)) x [1] dark red red dark red blue blue blue red blue Levels: red blue dark You have factor in data.frame, so you have to use levels(myDF$x) - ... Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] lmer and a response that is a proportion
Cameron Gillies cgillies at ualberta.ca writes: Hello Simon and John, I'm afraid I need to include random effects, both a random intercept and possibly random coefficients and it doesn't look like betareg can do that. Kevin Wright has posted wish on R-wiki for beta mixed effects model. There is no package for this, but there was a nice article describing such a model. Well, it is a start. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] showing available environment variables
Karin Lagesen karin.lagesen at medisin.uio.no writes: ... The default program for reading files when using openVignettes on my system is ggv. ggv is broken. I would therefore like to use a different program, and I found the variable R_PDFVIEWER which should do the trick. However, so far no luck. Therefore, how do I ask R what variables are available, and how do I get it to show what it is currently set to. Try options(pdfviewer=something) Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Fitting mixed-effects models with lme with fixed error term variances
Hello! Douglas Bates wrote: ... Not quite. There is now a capability in lmer to specify starting estimates for the relative precision matrices which means that you can use the estimates from one fit as starting estimates for another fit. It looks like fm1 - lmer(...) fm2 - lmer(y ~ x + (...), start = [EMAIL PROTECTED]) I should say that in my experience this has not been as effective as I had hoped it would be. What I see in the optimization is that the first few iterations reduce the deviance quite quickly but the majority of the time is spent refining the estimates near the optimum. So, for example, if it took 40 iterations to converge from the rather crude starting estimates calculated within lmer it might instead take 35 iterations if you give it good starting estimates. Yes, I also have the same experience with other programs, say VCE[1]. What I was hopping for was just solutions from linear mixed model i.e. either via GLS or PLS representations and no optimization if values for variance components are passed as input - there should be a way to distinguish that from just passing starting values.. The PLS representation in lmer is in terms of the relative variance-covariance matrices of the random effects where relative means relative to s^2. (In fact, the Omega slot contains the relative What exactly is s^2? precision matrices (inverses of the relative variance matrices) but its the same idea. All variance components are expressed relative to s^2.) If it is sufficient to fix these then you can easily do that. Just follow the code in lmer until it gets to .Call(mer_ECMEsteps, mer, cv$niterEM, cv$EMverbose) LMEoptimize(mer) - cv return(new(lmer, mer, and replace the first two lines by .Call(mer_factor, mer, PACKAGE = lme4) Hmm. Sounds simple. I will try this. Do you find any value in adding this as an option in lmer or perhaps new? function, say lmeSol() or something similar? A side-effect of performing the factorization is to evaluate the ML and REML deviances and store the result in the deviance slot. The follow the code in lmer part will get a bit tricky because of the number of functions that are hidden in the lme4 namespace but I'm sure you know how to get around that. Thank you very much! Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] low-variance warning in lmer
Ben Bolker bolker at zoo.ufl.edu writes: For block effects with small variance, lmer will sometimes ... My question is: what should I suggest to students when this situation comes up? Can anyone point me to appropriate references? (I haven't found anything relevant in Pinheiro and Bates, but I may not have looked in the right place ...) Gelman has quite some similar examples. Take a look at his schools example in the Bayesian book. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Fitting mixed-effects models with lme with fixed error term variances
Douglas Bates wrote: On 11/21/06, Gregor Gorjanc [EMAIL PROTECTED] wrote: Douglas Bates bates at stat.wisc.edu writes: ... Can you be more specific about which parameters you want to fix and which you want to vary in the optimization? It would be nice to have the ability to fix all variances i.e. variances of random effects. That gets tricky in terms of the parameterization of the variance-covariance matrices for vector-valued random effects. These matrices are not expressed in the conventional parameterization of variances and covariances or even variances and correlation because the conditions for the resulting matrix to be positive definite are not simple bounds or easily expressed transformations then the matrix is larger than 2 by 2. I suppose what I could do is to allow these matrices to be specified in the parameterization that is used in the optimization and provide a utility function to map from the conventional parameters to these. That would mean that you couldn't fix ,say, the variance of the intercept term for vector-valued random effects but allow the variance of a slope for the same grouping factor to be estimated. Well, you could but only in the fortune(Yoda) sense. Yes, I agree here. Thank you for the detailed answer. By the way, if you fix all the variances then what are you optimizing over? The fixed effects? In that case the solution can be calculated explicitly for a linear mixed model. The conditional estimates of the fixed effects given the variance components are the solution to a penalized linear least squares problem. (Yes, the solution can also be expressed as a generalized linear least squares problem but there are advantages to using the penalized least squares representation. Yup. It would really be great to be able to do that nicely in R, say use lmer() once and since this might take some time use estimates of variance components next time to get fixed and random effects. Is this possible with lmer or any related function - not in fortune(Yoda) sense ;) -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Fitting mixed-effects models with lme with fixed error term variances
Hello! Douglas Bates wrote: On 11/22/06, Gregor Gorjanc [EMAIL PROTECTED] wrote: Douglas Bates wrote: On 11/21/06, Gregor Gorjanc [EMAIL PROTECTED] wrote: Douglas Bates bates at stat.wisc.edu writes: ... Can you be more specific about which parameters you want to fix and which you want to vary in the optimization? It would be nice to have the ability to fix all variances i.e. variances of random effects. ... effects but allow the variance of a slope for the same grouping factor to be estimated. Well, you could but only in the fortune(Yoda) sense. Yes, I agree here. Thank you for the detailed answer. By the way, if you fix all the variances then what are you optimizing over? The fixed effects? In that case the solution can be calculated explicitly for a linear mixed model. The conditional estimates of the fixed effects given the variance components are the solution to a penalized linear least squares problem. (Yes, the solution can also be expressed as a generalized linear least squares problem but there are advantages to using the penalized least squares representation. Yup. It would really be great to be able to do that nicely in R, say use lmer() once and since this might take some time use estimates of variance components next time to get fixed and random effects. Is this possible with lmer or any related function - not in fortune(Yoda) sense ;) Not quite. There is now a capability in lmer to specify starting estimates for the relative precision matrices which means that you can use the estimates from one fit as starting estimates for another fit. It looks like fm1 - lmer(...) fm2 - lmer(y ~ x + (...), start = [EMAIL PROTECTED]) I should say that in my experience this has not been as effective as I had hoped it would be. What I see in the optimization is that the first few iterations reduce the deviance quite quickly but the majority of the time is spent refining the estimates near the optimum. So, for example, if it took 40 iterations to converge from the rather crude starting estimates calculated within lmer it might instead take 35 iterations if you give it good starting estimates. Yes, I also have the same experience with other programs, say VCE[1]. What I was hopping for was just solutions from linear mixed model i.e. either via GLS or PLS representations and no optimization if values for variance components are passed as input - there should be a way to distinguish that from just passing starting values.. [1]http://vce.tzv.fal.de/index.pl Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] multiple R listservs?
Jeff Miller jeffmiller at adsam.com writes: I thought more about this last night after my email about putting an end to Read the Manual replies. ... A potential solution is to offer two lists.one for newbies and one for.umm.not-newbies. Some not-newbies may prefer subscribing to both and setting up 2 email folders. There is R-help and R-devel already. If you find to much traffic on the R-help you could read only those mails that attract you - that is what I do. Additionally, using Gmane with Bloglines is very very nice for my inbox. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] query in R
Jin, Xiaodong Jin close2ceo at yahoo.com writes: how to realize the following SQL command in R? select distinct A, B, count(C) from TABLE group by A, B ; quit; if you want to perform such query on a data.frame! then table(table$a, table$b) will produce similar output, but in 2-way table, or this for similar output table(paste(table$a, -, table$b)) Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Probit Scale
nelson - nelson1977 at gmail.com writes: Hi all! i search but i found no info, so i'm heere to ask you! How can i plot a graph using the probit scale on y axes? Not that I know, however you can transform y values via pnorm() i.e. plot(x=x, y=pnorm(q=y, ...)) Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Composition like classes functions
Gregor Gorjanc gregor.gorjanc at bfro.uni-lj.si writes: Hello! I am working on breed composition of animals in my project. Say I have an animal 1 with parents 2 (father) and 3 (mother). If father is of breed A and mother of breed B, then their descendant is of breed AB or 50 % of A and 50 % of B. I would like to have a general way to represent this kind of data. I was thinking about list structure i.e. ... However, I would like to ask if there is anything like this already in R or contributed packages. Replying myself. There is a package compositions at CRAN. I should have done homework before asking here. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Placing of legends
Thevini Christian Lerch t.c.l at gmx.net writes: placings of legends is sometimes tricky. For placing outside the plot region I found locator to be useful. Unfortunately, the click defines the upper left corner. Is there a way to change this corner (say lower right corner)? Not sure about outside plot region but you can specify position very easy. Take a look at legend help page - 3rd paragraph under details. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Date, date, POSIX question
Joe W. Byers Joe-Byers at utulsa.edu writes: ... Can anyone explain the benefits and detractions of the 'Date' package verses the 'date' package and verses 'POSIX' dates. ... Not sure which one should be prefered, but there was an article in Rnews that nicely covered pros and cons. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] read.fwf and header
Greg Snow wrote: How about using a connection and reading the header separate from the data, like this: tmp1 - file('c:/temp/tmp.dat') open(tmp1) my.names - scan(tmp1, nlines=1, what='') new.data-read.fwf(file=tmp1, widths=c(3, 4, 10, 3, 2, 2, 2, 2, 11, 19), header=FALSE) names(new.data) - my.names close(tmp1) Yes, also possible as has been shown in previous posts. -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] read.fwf and header
Martin Maechler wrote: Gregor == Gregor Gorjanc [EMAIL PROTECTED] on Mon, 30 Oct 2006 23:33:21 +0100 writes: Gregor Daniel Nordlund wrote: Gregor, According to the help for read.fwf, sep needs to be set to a value that occurs only in the header record. I changed the spaces to commas in the header record of your example and used the following syntax and was able to read the file just fine. new.data-read.fwf(file=test.txt, widths=c(3, 4, 10, 3, 2, 2, 2, 2, 11, 19), header=TRUE, sep=',') Hope this is helpful, Dan Gregor Thanks Dan! But I have to modfy file first. Not that Gregor much of work but still. Yes, but I think it shows read.fwf() should not be extended for even more special cases: In my (and probably R-core's) view, read.fwf() should only have to be used for ``legacy data files'' (those times when people used *no* separators in order to save disk space), since nowadays, such data files should automatically have correct separators. -- Fix the file producing process rather than make read.fwf() unnecessarily more complicated. Thank you for this explanation of your (and probably R-core's) view! I really appreciate such feedback. I do agree that read.fwf is a bit archaic way to import data, but sometimes you can not fix file producing process. Perhaps above explanation and code examples from this thread could be added to read.fwf help page. I can provide a patch if my proposal is sane. Regards, Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] Composition like classes functions
Hello! I am working on breed composition of animals in my project. Say I have an animal 1 with parents 2 (father) and 3 (mother). If father is of breed A and mother of breed B, then their descendant is of breed AB or 50 % of A and 50 % of B. I would like to have a general way to represent this kind of data. I was thinking about list structure i.e. list(A=0.5, B=0.5) $A [1] 0.5 $B [1] 0.5 Additionally, it should accommodate for more than 2 classes. However, I would like to ask if there is anything like this already in R or contributed packages. Thanks! -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] read.fwf and header
Hi! I have data (also in attached file) in the following form: num1 num2 num3 int1 fac1 fac2 cha1 cha2 Date POSIXt 11 f q 1900-01-01 1900-01-01 01:01:01 2 1.0 131.5 2 a g r z1900-01-01 01:01:01 3 1.5 1188830.5 3 b h s y 1900-01-01 1900-01-01 01:01:01 4 2.0 1271846.3 4 c i t x 1900-01-01 1900-01-01 01:01:01 5 2.5 829737.4d j u w 1900-01-01 6 3.0 1240967.3 5 e k v v 1900-01-01 1900-01-01 01:01:01 7 3.5 919684.4 6 f l w u 1900-01-01 1900-01-01 01:01:01 8 4.0 968214.6 7 g m x t 1900-01-01 1900-01-01 01:01:01 9 4.5 1232076.4 8 h n y s 1900-01-01 1900-01-01 01:01:01 10 5.0 1141273.4 9 i o z r 1900-01-01 1900-01-01 01:01:01 5.5 988481.4 10 j q 1900-01-01 1900-01-01 01:01:01 This is a FWF (fixed width format) file. I can not use read.table here, because of missing values. I have tried with the following read.fwf(file=test.txt, widths=c(3, 4, 10, 3, 2, 2, 2, 2, 11, 20), header=TRUE) Error in read.table(file = FILE, header = header, sep = sep, as.is = as.is, : more columns than column names I could use: read.fwf(file=test.txt, widths=c(3, 4, 10, 3, 2, 2, 2, 2, 11, 20), header=FALSE, skip=1) V1 V2V3 V4 V5 V6 V7 V8 V9 V10 1 1 NANA 1f q 1900-01-01 1900-01-01 01:01:01 2 2 1.0 131.5 2 a g r z 1900-01-01 01:01:01 3 3 1.5 1188830.5 3 b h s y 1900-01-01 1900-01-01 01:01:01 4 4 2.0 1271846.3 4 c i t x 1900-01-01 1900-01-01 01:01:01 5 5 2.5 829737.4 NA d j u w 1900-01-01 6 6 3.0 1240967.3 5 e k v v 1900-01-01 1900-01-01 01:01:01 7 7 3.5 919684.4 6 f l w u 1900-01-01 1900-01-01 01:01:01 8 8 4.0 968214.6 7 g m x t 1900-01-01 1900-01-01 01:01:01 9 9 4.5 1232076.4 8 h n y s 1900-01-01 1900-01-01 01:01:01 10 10 5.0 1141273.4 9 i o z r 1900-01-01 1900-01-01 01:01:01 11 NA 5.5 988481.4 10 jq 1900-01-01 1900-01-01 01:01:01 Does anyone have a clue, how to get above result with header? Thanks! -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] read.fwf and header
Marc Schwartz wrote: On Mon, 2006-10-30 at 19:51 +0100, Gregor Gorjanc wrote: Hi! I have data (also in attached file) in the following form: num1 num2 num3 int1 fac1 fac2 cha1 cha2 Date POSIXt 11 f q 1900-01-01 1900-01-01 01:01:01 2 1.0 131.5 2 a g r z1900-01-01 01:01:01 3 1.5 1188830.5 3 b h s y 1900-01-01 1900-01-01 01:01:01 4 2.0 1271846.3 4 c i t x 1900-01-01 1900-01-01 01:01:01 5 2.5 829737.4d j u w 1900-01-01 6 3.0 1240967.3 5 e k v v 1900-01-01 1900-01-01 01:01:01 7 3.5 919684.4 6 f l w u 1900-01-01 1900-01-01 01:01:01 8 4.0 968214.6 7 g m x t 1900-01-01 1900-01-01 01:01:01 9 4.5 1232076.4 8 h n y s 1900-01-01 1900-01-01 01:01:01 10 5.0 1141273.4 9 i o z r 1900-01-01 1900-01-01 01:01:01 5.5 988481.4 10 j q 1900-01-01 1900-01-01 01:01:01 This is a FWF (fixed width format) file. I can not use read.table here, because of missing values. I have tried with the following read.fwf(file=test.txt, widths=c(3, 4, 10, 3, 2, 2, 2, 2, 11, 20), header=TRUE) Error in read.table(file = FILE, header = header, sep = sep, as.is = as.is, : more columns than column names I could use: read.fwf(file=test.txt, widths=c(3, 4, 10, 3, 2, 2, 2, 2, 11, 20), header=FALSE, skip=1) V1 V2V3 V4 V5 V6 V7 V8 V9 V10 1 1 NANA 1f q 1900-01-01 1900-01-01 01:01:01 2 2 1.0 131.5 2 a g r z 1900-01-01 01:01:01 3 3 1.5 1188830.5 3 b h s y 1900-01-01 1900-01-01 01:01:01 4 4 2.0 1271846.3 4 c i t x 1900-01-01 1900-01-01 01:01:01 5 5 2.5 829737.4 NA d j u w 1900-01-01 6 6 3.0 1240967.3 5 e k v v 1900-01-01 1900-01-01 01:01:01 7 7 3.5 919684.4 6 f l w u 1900-01-01 1900-01-01 01:01:01 8 8 4.0 968214.6 7 g m x t 1900-01-01 1900-01-01 01:01:01 9 9 4.5 1232076.4 8 h n y s 1900-01-01 1900-01-01 01:01:01 10 10 5.0 1141273.4 9 i o z r 1900-01-01 1900-01-01 01:01:01 11 NA 5.5 988481.4 10 jq 1900-01-01 1900-01-01 01:01:01 Does anyone have a clue, how to get above result with header? Thanks! The attachment did not come through. Perhaps it was too large? Not sure if this is the most efficient way, but how about this: DF - read.fwf(test.txt, widths=c(3, 4, 10, 3, 2, 2, 2, 2, 11, 20), skip = 1, strip.white = TRUE, col.names = read.table(test.txt, nrow = 1, as.is = TRUE)[1, ]) Argh, my fault as I forgot to attach it :( Not sure if this is the most efficient way, but how about this: DF - read.fwf(test.txt, widths=c(3, 4, 10, 3, 2, 2, 2, 2, 11, 20), skip = 1, strip.white = TRUE, col.names = read.table(test.txt, nrow = 1, as.is = TRUE)[1, ]) That is a very nice compromise! No need for [1, ], due to nrow=1. Of course, with the limited number of columns, you can always just set colnames(DF) - c(num1, num2, num3, int1, fac1, fac2, cha1, cha2, Date, POSIXt) I fully agree here, but I kind of lack this directly in read.fwf. I hope that someone from R-core is also listening to this ;) Thank you! Gregor num1 num2 num3 int1 fac1 fac2 cha1 cha2 Date POSIXt 11 f q 1900-01-01 1900-01-01 01:01:01 2 1.0 131.5 2 a g r z1900-01-01 01:01:01 3 1.5 1188830.5 3 b h s y 1900-01-01 1900-01-01 01:01:01 4 2.0 1271846.3 4 c i t x 1900-01-01 1900-01-01 01:01:01 5 2.5 829737.4d j u w 1900-01-01 6 3.0 1240967.3 5 e k v v 1900-01-01 1900-01-01 01:01:01 7 3.5 919684.4 6 f l w u 1900-01-01 1900-01-01 01:01:01 8 4.0 968214.6 7 g m x t 1900-01-01 1900-01-01 01:01:01 9 4.5 1232076.4 8 h n y s 1900-01-01 1900-01-01 01:01:01 10 5.0 1141273.4 9 i o z r 1900-01-01 1900-01-01 01:01:01 5.5 988481.4 10 j q 1900-01-01 1900-01-01 01:01:01 __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] read.fwf and header
Daniel Nordlund wrote: Gregor, According to the help for read.fwf, sep needs to be set to a value that occurs only in the header record. I changed the spaces to commas in the header record of your example and used the following syntax and was able to read the file just fine. new.data-read.fwf(file=test.txt, widths=c(3, 4, 10, 3, 2, 2, 2, 2, 11, 19), header=TRUE, sep=',') Hope this is helpful, Dan Thanks Dan! But I have to modfy file first. Not that much of work but still. Regards, Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Header of dataframe
Benjamin Dickgiesser dickgiesser at gmail.com writes: ... Is there an easy way to add a column name to the first column? What about summary.aggregate - function(y, ...) { temp.mean - aggregate(y, FUN=mean, ...) temp.sd - aggregate(y, FUN=sd, ...) temp.length - aggregate(y, FUN=length, ...) temp - data.frame(cbind(col=names(y), mean=temp.mean$x, stdev=temp.sd$x,n=temp.length$x)) } or summary.aggregate - function(y, ...) { temp.mean - aggregate(y, FUN=mean, ...) temp.sd - aggregate(y, FUN=sd, ...) temp.length - aggregate(y, FUN=length, ...) temp - data.frame(cbind(mean=temp.mean$x, stdev=temp.sd$x,n=temp.length$x)) rownames(temp) - names(y) } Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Mixed conditional logit model
Ryuichi Tamura ry.tamura at gmail.com writes: I wrote a collection of R functions for estimating discrete choice models by simulated maximum likelihood. It includes: ... My R functions is based on GAUSS program listed in this site. Have you considered to wrap all funcs in a package? Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] Problem reading non-ISO data via read.csv2
Hello! I have a CSV file and when I try to import it into R with read.csv2 I get the following error: R read.csv2(file=pasme2.csv, na.strings=) Error in type.convert(data[[i]], as.is = as.is[i], dec = dec, na.strings = character(0)) : invalid multibyte string There are some strange characters in it, but I never experienced such behaviour. Actually, this file was produced with R (2.4.0) on Windows! Utility file under Linux says: $ file pasme2.csv pasme.csv: Non-ISO extended-ASCII text I am attaching few lines of a file for example. And mandatory info: R version 2.4.0 (2006-10-03) i486-pc-linux-gnu locale: LC_CTYPE=en_GB.UTF-8;LC_NUMERIC=C;LC_TIME=en_GB.UTF-8;LC_COLLATE=en_GB.UTF-8;LC_MONETARY=en_GB.UTF-8;LC_MESSAGES=en_GB.UTF-8;LC_PAPER=en_GB.UTF-8;LC_NAME=C;LC_ADDRESS=C;LC_TELEPHONE=C;LC_MEASUREMENT=en_GB.UTF-8;LC_IDENTIFICATION=C -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. -- __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Problem reading non-ISO data via read.csv2
Prof Brian Ripley wrote: ... There are some strange characters in it, but I never experienced such behaviour. Actually, this file was produced with R (2.4.0) on Windows! Well, then it cannot be a UTF-8 file (there are no UTF-8 locales on Windows, not any means to write a UTF-8 file here), and you have told R to read it in your UTF-8 locale. You need to specify the correct encoding: see ?file. But then you would have to do that in any application with such a text file, as an application could at best guess the encoding. Thank you very much. Your suggestion solved my problem i.e. I have used: read.csv2(file=file(pasme2.csv, encoding=WINDOWS-1252)) Utility file under Linux says: $ file pasme2.csv pasme.csv: Non-ISO extended-ASCII text I am attaching few lines of a file for example. And mandatory info: No attached file appeared. That is weird as my sent folder shows mail with an attachment. Anyway, problem is solved now - with Your help! -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] [Fwd: [AGDG-LIST:405] R Computing Contest]
Original Message Subject: [AGDG-LIST:405] R Computing Contest Date: Sat, 21 Oct 2006 12:08:13 -0400 From: Larry Schaeffer [EMAIL PROTECTED] Reply-To: [EMAIL PROTECTED] To: Animal Geneticist's Discussion [EMAIL PROTECTED] For those that are interested only: R Computer Programming Challenge Given: y = Factor A + Factor B + b1(Covariate1) + b2(Covariate2) + ... + bp(Covariate p) + animal + e where y is a single trait - vector of observations, Factors A and B are fixed with ma and mb levels respectively, Covariates 1 to p are fixed regressions and p is a general number but less than 20. Heritability is a variable and should be requested by the program. You are given a list of the animal IDs of animals with records, and their sires and dams. Sire and dam IDs are smaller than the IDs of their progeny, and all IDs go from 1 to n where n can be up to 10 million animals. Tasks: 1) Estimate the EBVs of all animals using BLUP and mixed model equations and list the top 10 largest EBVs with associated animal IDs, first ten levels of Factor A with solutions, first ten levels of Factor B with solutions, and all estimated regression coefficients. Submit your R program to Bill Szkotnicki ([EMAIL PROTECTED]) by January 5, 2007. Submissions will be dated by their arrival date. Provide any special functions or details that might be needed to run your program. All R programs will be run on the same test data sets, computer (Linux), and timed. Winner: Earliest submitted R program that runs the fastest and gives the correct answers. Prize: Free registration for the Kennedy Conference in 2008 in Guelph. -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] object(s) are masked from package - what does it mean?
tom soyer tom.soyer at gmail.com writes: Hi, Sometime when I attach a dataset, R gives me the following message/warning:The following object(s) are masked from package:datasets: column_name. Does anyone know what it means? Since it seems that the dataset was attached and I could manipulate the data from the dataset without problems, I am wondering what was R trying to tell me. As far as I understand it: Imagine there is a package A with object tmp and you load it. When you load another package B with object of the same name i.e. tmp then you will access tmp from package B and not A. You might work though with namespaces, :: and ::: Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] glm function question
Chris Linton connect.chris at gmail.com writes: I am creating a model attempting to predict the probability someone will reoffend after being caught for a crime. There are seven total inputs and I planned on using a logistic regression. I started with a null deviance of 182.91 and ended up with a residual deviance of 83.40 after accounting for different interactions and such. However, I realized after that my code is different from that in my book. And I can't figure out what I need to put in it's place. Here's my code: ... fit1h = glm(reoff ~ factor(subst) + factor(violence) + prior + factor(violence):factor(subst) + factor(violence):factor(educ) + factor(violence):factor(age) + factor(violence):factor(prior)) summary(fit1h) If you noticed, there's no part of my code that looks like: family=binomial(link=logit)) ... However, when I do this, my null deviance is 1104 and my residual deviance is 23460. THIS IS A HUGE DIFFERENCE IN MODEL FIT! I'm not sure if I have to redo my model or if my book was simply doing the family=binomial(link=logit) for a specific problem/reason. You state that you model the *probability* that ... Then family=gaussian, which is the default data generation model in glm is not appropriate. Yes, you need to use family=binomial(link=logit) or family=binomial(link=probit), but you also need to take care in proper specification of your y in the glm call. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] winbugs - R question
allan clark allan at stats.uct.ac.za writes: hi all i would like to use r and winbugs in order to undertake seemingly unrelated regression. i am using the R2WINBUGS library. i just simulated a simple example (sample size is 25) in order to get the correct code. i suspect that the problem is in my definition of the prior. it wants a multivariate node. No solution, but just a note that this is more a question for (Win)BUGS list. It is not very active recently, though. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Transcript of Conservative ANOVA tables
Dieter Menne dieter.menne at menne-biomed.de writes: Dear friends of lmer, http://wiki.r-project.org/rwiki/doku.php?id=guides:lmer-tests I have put a transcript of the long thread on lmer/lme4 statistical test into the Wiki. For all those who missed it life, and for those like me, who don't like the special style of the R-list to keep full length quotes. Creating the text there was not much fun, waiting times are simply unacceptable and the Wiki only give an empty page when syntax errors (for example from quotes) are detected. I agree that there is a problem about waiting times with large pages. This might be of interest for r-sig-wiki list, but I can not CC from Gmane - I will send separate mail. I think there was discussion about this and that wiki is optimized for many small pages. I do agree though that having several pages for this transcript is not acceptable. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Sweave, latex Warning
Göran Broström goran.brostrom at gmail.com writes: When I am using Sweave to produce reports I get a strange Warning from latex: LaTeX Warning: You have requested package `/usr/local/lib/R/share/texmf/Sweave' but the package provides `Sweave'. It seems to be completely harmless, but I wonder if it is a symptom of installation problems, R or latex. How can I get rid of it? This is on linux (ubuntu), with R-2.4.0. What do you actually use in Rnw file? Is it \usepackage{/usr/local/lib/R/share/texmf/Sweave} or \usepackage{Sweave} I think this was discussed in r-devel list a while ago. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] sweave2html only for Linux users ?
Laurent Rhelp laurentRhelp at free.fr writes: Dear R-List, I would like to convert my rnw files written with Sweave from LaTeX to html. I discovered on the list the following link : http://biostat.mc.vanderbilt.edu/twiki/bin/view/Main/SweaveConvert Is this document only for linux users or I can use it with windose XP ? Maybe cygwin can be of help here. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] mcmcsamp-CI instead of P-values - references?
Henrik Parn henrik.parn at bio.ntnu.no writes: Dear all, Given the discussions and issues of d.f., p-values and mcmcsamp-CIs in mixed models, I wonder if anyone could recommend one or two papers (or other citable sources for that sake) that summarizes the arguments for/against P-values/mcCIs. I have followed the discussions on R-help and have learnt a lot (in my naive non-statistician way). With all respect to all great contributions, I may need published articles (if there are any), that could be suitable to use in the Methods-section of a manuscript, to motivate a no-p-values-but-some-nice-mcmcsampCIs-instead-approach! Book (Bayesian Data Analysis) by Gelman et al. is very good source of this and much more. You might also find his papers and blog interesting http://www.stat.columbia.edu/~gelman/ http://www.stat.columbia.edu/~gelman/blog/ Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] how to check object size in workspace
Weiwei Shi helprhelp at gmail.com writes: hi, I tried to keep my workspace as small as possible when i tried to save.image(). Sometimes, I am not sure whether obj in workspace will be needed in future or not. However, I want to delete some big ones when I use save.image(). BTW, I know I could use save, but I would like to know : how to check objects' size (the size when I try to save it in disk) in workspace, anyway. Take a look at ?object.size and also in ?ll in gdata package Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Adding graphics to R-help Files
John James jjbarn at btinternet.com writes: Is there any way to add graphics (e.g. thumbnails in examples) to Rd files? A picture is worth a thousand words! No. You can provide code example that works. That is quite close. You might also look at http://bg9.imslab.co.jp/Rhelp/ http://addictedtor.free.fr/graphiques/ Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Creating Movies with R
Hello! J.R. Lockwood lockwood at rand.org writes: An alternative that I've used a few times is the jpg() function to create the sequence of images, and then converting these to an mpeg movie using mencoder distributed with mplayer. This works on both windows and linux. I have a pretty self-contained example file written up that I can send to anyone who is interested. Oddly, the most challenging part was creating a sequence of file names that would be correctly ordered - for this I use: lex - function(N){ ## produce vector of N lexicograpically ordered strings ndig - nchar(N) substr(formatC((1:N)/10^ndig,digits=ndig,format=f),3,1000) } RWiki[1] would be a very nice place for such explanation. I am looking forwrad to it! Gregor [1]http://wiki.r-project.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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Conservative
Douglas Bates bates at stat.wisc.edu writes: On 9/13/06, Dimitris Rizopoulos dimitris.rizopoulos at med.kuleuven.be I believe that the LRT is anti-conservative for fixed effects, as described in Pinheiro and Bates companion book to NLME. You have this effect if you're using REML, for ML I don't think there is any problem to use LRT between nested models with different fixed-effects structure. ... The other question is how does one evaluate the likelihood-ratio test statistic and that is the issue that Dimitris is addressing. The REML criterion is a modified likelihood and it is inappropriate to look at differences in the REML criterion when the models being compared have different fixed-effects specifications, or even a different parameterization of the fixed effects. However, the anova method for an lmer object does not use the REML criterion even when the model has been estimated by REML. It uses the profiled log-likelihood evaluated at the REML estimates of the relative variances of the random effects. That's a complicated statement so let me break it down. ... Is this then the same answer as given by Robinson:1991 (ref at the end) to question by Robin Thompson on which likelihood (ML or REML) should be used in testing the fixed effects. Robinson answered (page 49 near bottom right) that both likelihoods give the same conclusion about fixed effects. Can anyone comment on this issues? Thanks, Gregor @Article{Robinson:1991, author = {Robinson, G. K.}, title ={That {BLUP} is a good thing: the estimation of random effects}, journal = ss, year = {1991}, volume = {6}, number = {1}, pages ={15--51}, keywords = {BLUP, example, derivations, links, applications}, vnos = {GG} } __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
Prof Brian Ripley ripley at stats.ox.ac.uk writes: ... Check out tests/internet.R. nsl() checks if you can resolve host names, which has worked well enough there. Thank you prof. Ripley for this pointer. I am posting here the relevant part if someone does not look at SVN. I would just like to ask why is .Platform$OS.type == unix added to the test? Is nsl() available only on unix like platforms or ... I did not found any specifics in its help page. if(!capabilities()[http/ftp]) { warning(no internet capabilities) q() } if(.Platform$OS.type == unix is.null(nsl(cran.r-project.org))) q() Does it make any sense to write a function that would use these two tests. isNetAvailable - function() { ifelse(!capabilities()[http/ftp] ## .Platform$OS.type == unix ## ??? is.null(nsl(cran.r-project.org)), FALSE, TRUE) } Regards, Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
Gabor Grothendieck wrote: Here is a variation for Windows. The second line returns TRUE or FALSE and may need to be varied if the output of ping is not the same on your system as on mine: ping - system(ping www.google.com, intern = TRUE) as.numeric(strsplit(grep(Received, ping, value = TRUE), [ ,])[[1]][8]) 0 ... Thank you Gabor and Ted! These are all fine ways, but as both of you have stated not really general as you never know what ping will produce on different versions etc. I am really keen on test from tests/internet.R. If that test is OK for base R, I do not see why it should not be OK for R package. I just hope to get some more feedback on my question[1] to prof. Ripley. [1]https://stat.ethz.ch/pipermail/r-help/2006-September/112766.html Thank you! -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
Prof Brian Ripley wrote: On Mon, 11 Sep 2006, Gregor Gorjanc wrote: Prof Brian Ripley ripley at stats.ox.ac.uk writes: ... Check out tests/internet.R. nsl() checks if you can resolve host names, which has worked well enough there. Thank you prof. Ripley for this pointer. I am posting here the relevant part if someone does not look at SVN. I would just like to ask why is .Platform$OS.type == unix added to the test? Is nsl() available only on unix like platforms or ... I did not found any specifics in its help page. Did you look at the help page on Windows? Looking at the help page on Unix only tells you about Unix. Hint: the help page is src/library/utils/man/unix/nsl.Rd (In my country, PhD students are supposed to be able to find things like that out for themselves.) ... Thank you for additional pointer. I did not look windows help page as I do not have windows at disposal all the time, but You are right - I should have looked into the sources. I found out that there is no such function i.e. nsl() under windows. If I summarize the thread there is (currently) no way to test for internet presence with a general approach. -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Matrix multiplication using apply() or lappy() ?
Jeffrey J. Hallman jhallman at frb.gov writes: Tim Hesterberg timh at insightful.com writes: toby_marks at americancentury.com asked: I am trying to divide the columns of a matrix by the first row in the matrix. Dividing columns of a matrix by a vector is a pretty fundamental operation, and the query resulted in a large number of suggestions: It is unsatisfactory when such a fundamental operation is done in so many different ways. * It makes it hard to read other people's code. * Some of these are very inefficient. But since you have no way to force people to use your new 'standard' operators, people can and will still use any of those myriad ways you decry to do the same thing. It's part of the price you pay for working with a flexible system. Besides, nothing will ever make it easy to read other people's code. Reading others code is never an easy task, but there are huge differences between various authors and I agree with Tim, that it is nice to have some standardized operators for common/fundamental operations. If there are standard operators, people will use it more an more. __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] Test internet presence
Hello useRs! I have a small package and I need internet access for examples. Of course it works fine when I have internet access, but not otherwise. I remember I saw a way to test availability of internet access, but I can not find it now in archives. Can anyone here help me? Thanks! -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
David Barron wrote: I means non-zero, not non-negative. ... Thanks. download.file() is fine! -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
Gregor Gorjanc wrote: Hello useRs! I have a small package and I need internet access for examples. Of course it works fine when I have internet access, but not otherwise. I remember I saw a way to test availability of internet access, but I can not find it now in archives. Can anyone here help me? David Barron wrote: I means non-zero, not non-negative. ... Thanks. download.file() is fine! Thinking a bit more, download.file() is not really general as I need to specify an url and it might happen that I have internet connection, but server that provides url is down. I am to picky on this? -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
Gregor Gorjanc wrote: Gregor Gorjanc wrote: Hello useRs! I have a small package and I need internet access for examples. Of course it works fine when I have internet access, but not otherwise. I remember I saw a way to test availability of internet access, but I can not find it now in archives. Can anyone here help me? David Barron wrote: I means non-zero, not non-negative. ... Thanks. download.file() is fine! Thinking a bit more, download.file() is not really general as I need to specify an url and it might happen that I have internet connection, but server that provides url is down. I am to picky on this? Additionally, download.file throws an error if url is not available i.e. if I do not have internet access. I have used the following example: if(download.file(url=http://www.r-project.org/index.html;, destfile=tempfile()) == 0) { print(Yuhuu) } unpluged the net cable and got the following message during R CMD check (in examples part) trying URL 'http://www.r-project.org/index.html' Warning: unable to resolve 'www.r-project.org'. Error in download.file(url = http://www.r-project.org/index.html;, destfile = tempfile()) : cannot open URL 'http://www.r-project.org/index.html' Execution halted It seems that 'internal' method was used (I use R 2.3.1 under Linux) as indicated in help page of download.file. I could use wget or lynx methods, but these two must be available, so this is not really portable. Are there any other options for testing internet access? I am thinking that this might be more relevant for R-devel. I will wait a bit before moving there. -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Comparison of R with proprietary software
arin basu arinbasu at cashette.com writes: Hi All: Recently, I got a query how R compared cost-wise (cost of purchase, maintenance, and training manuals) with other proprietary statistical software programs (or programming environments). The request, verbatim: Take a look at http://wiki.r-project.org/rwiki/doku.php?id=getting-started:translations:translations and links from there! Regards, Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
Thank you. Your solution is usable but unfortunatelly not portable to Windows. I would like to use this test in package check, which can include also windows OS. On 11-Sep-06 Ted Harding wrote: On 11-Sep-06 Gregor Gorjanc wrote: It seems that 'internal' method was used (I use R 2.3.1 under Linux) as indicated in help page of download.file. I could use wget or lynx methods, but these two must be available, so this is not really portable. Are there any other options for testing internet access? I am thinking that this might be more relevant for R-devel. I will wait a bit before moving there. -- Lep pozdrav / With regards, Gregor Gorjanc Hi Gregor, Since you are using Linux, I think you should ask R to delegate the test to the system. If you have a script, in executable file (755) say test.inet.sh, which says something like if ping -c 1 something ; then export NET_UP=YES else export NETP_UP=NO fi where something is the IP address or name of an external host which responds to 'ping' (some will not, depending on their firewall settings), then you can use on R: system(test.inet) if( sys.getenv(NET_UP) == YES ) { ... } else { ... } system(test.inet.sh) if( sys.getenv(NET_UP) == YES ) { ... } else { ... } For example (nothing to do with R, but shows the principle), I have the following script to set my system time and hardware clock from whichever one of 3 NTP servers is willing to respond: if /bin/ping -c 1 ntp0.zen.co.uk ; then export NETTIME=/usr/sbin/ntpdate -u ntp0.zen.co.uk elif /bin/ping -c 1 ntp2b.mcc.ac.uk ; then export NETTIME=/usr/sbin/ntpdate -u ntp2b.mcc.ac.uk elif /bin/ping -c 1 ntp2c.mcc.ac.uk ; then export NETTIME=/usr/sbin/ntpdate -u ntp2c.mcc.ac.uk else export NETTIME= fi if [ $NETTIME != ] ; then sleep 1 sleep 1 $NETTIME /sbin/clock -u -w date fi which also illustrates how to allow for the possibility that the default server might not be responding at the time, so it has 2 fallback servers. -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] R drop behavior -- set as option in later version?
Balaji S. Srinivasan balajis at stanford.edu writes: ... Hello, I agree with you that you find yourself typing the same constructs over and over. I think that we need to distinguish two modes of working with R. If you do analysis, then you can really get tired of typing drop=FALSE, na.rm=TRUE etc. But this things are very important when you are programming in R. Since these two modes are not really separated in R I do not think this is an easy task, but it would be great to have it. I had recently the same question for na.rm=TRUE. But imagine how hard would it be to have two separate modes ... argh, probably a mess^2 or have I missed something obvious. The possible caveats are: 1) Performance issues associated with GetOption call (this is probably not a showstopper) 2) Breaking old code which assumes drop=TRUE if the drop=FALSE option is set. For the latter situation, it would be possible to wrap any new code with an options setting as follows: og - options(GlobalDrop) options(GlobalDrop=FALSE) #new code here which assumes drop=FALSE options(og) I think that local options i.e. drop=TRUE in code should override global ones. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] R drop behavior -- set as option in later version?
Martin Maechler wrote: Wrong Mailing List !! Proposals for changes to R should be discussed on R-devel, see the posting guide. I'll reply separately, but only CC to R-devel. You are right, but I think that this might be an overkill. But, who am I to decide that anyway! I actually excited that you noticed this thread and find it usefull for r-devel discussion. -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] Producing R demos
Hello! I have found terrific demo or R package functionality at http://had.co.nz/reshape/french-fries-demo.html Author has told me off-list that he is using SnapzProX (on mac), and ghostwriter (http://had.co.nz/ghostwriter/) to automate the typing. Unfortunatelly, I do not have mac ;) Can anyone on the list suggest, how such a demo (video + automated typing of a script) could be produced under either Windows or Linux OS? I would like to create such a demo for presentation as I will not be able to install R on the machine, but I could play a movie. Thanks! -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Producing R demos
Gabor Grothendieck wrote: I have not used this but wink is a free demo-maker for Windows available at: http://www.debugmode.com/wink/ Click the red on green arrow on that page to a (very short) demo. Thanks! Do you perhaps also know equivalent of ghostwriter (link is bellow) for windows/linux? On 8/29/06, Gregor Gorjanc [EMAIL PROTECTED] wrote: Hello! I have found terrific demo or R package functionality at http://had.co.nz/reshape/french-fries-demo.html Author has told me off-list that he is using SnapzProX (on mac), and ghostwriter (http://had.co.nz/ghostwriter/) to automate the typing. Unfortunatelly, I do not have mac ;) Can anyone on the list suggest, how such a demo (video + automated typing of a script) could be produced under either Windows or Linux OS? I would like to create such a demo for presentation as I will not be able to install R on the machine, but I could play a movie. Thanks! -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] day, month, year functions
Gabor Grothendieck wrote: On 8/17/06, Martin Maechler [EMAIL PROTECTED] wrote: Gregor == Gregor Gorjanc [EMAIL PROTECTED] on Fri, 11 Aug 2006 00:27:27 + (UTC) writes: [snip] There are two problems: 1. as.POSIXlt is not generic. (This problem may not be too important given that as.POSIXlt does handle Date and chron dates classes already but in terms of handling all potential classes its a limitation.) But as.POSIXlt can be implemented as generic. Given the variatey of potential classes this would be a reasonable move. -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Setting contrasts for polr() to get same result of SAS
T Mu muster at gmail.com writes: Hi all, I am trying to do a ordered probit regression using polr(), replicating a result from SAS. polr(y ~ x, dat, method='probit') suppose the model is y ~ x, where y is a factor with 3 levels and x is a factor with 5 levels, To get coefficients, SAS by default use the last level as reference, R by default use the first level (correct me if I was wrong), Yes. I tried relevel, reorder, contrasts, but no success. I found what really matters is I am sure those can help but you need to be carefull to reorder levels that the order is the same in SAS and R. options(contrasts = c(contr.treatment, contr.poly)) or options(contrasts = c(contr.SAS, contr.poly)) You can also set contrasts directly to factors via contrasts(x) - contr.SAS where x is your factor. You can also set different contrasts to different factors. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] day, month, year functions
Martin Maechler wrote: Gregor == Gregor Gorjanc [EMAIL PROTECTED] on Fri, 11 Aug 2006 00:27:27 + (UTC) writes: Gregor Gabor Grothendieck ggrothendieck at gmail.com writes: Here are three ways: xx - as.Date(2006-01-05) # 1. use as.POSIXlt as.POSIXlt(xx)$mday as.POSIXlt(xx)$mon + 1 as.POSIXlt(xx)$year + 1900 # 2. use format as.numeric(format(xx, %d)) as.numeric(format(xx, %m)) as.numeric(format(xx, %Y)) # 3. use month.day.year in chron package library(chron) month.day.year(unclass(xx))$day month.day.year(unclass(xx))$month month.day.year(unclass(xx))$year Gregor Hi, Gregor it would really be great if there would be Gregor sec(), min(), hour() day(), month(), year() Gregor generic functions that would work on all date classes. Where Gregor applicable of course. I imagine that argument to get out integer Gregor or character would alse be nice. I disagree pretty strongly: - We definitely don't want min() to return minutes instead of minimum ! Pheu, a good catch. You are definitely right! - Why pollute the namespace with 6 (well, actualy 5!) new function names, when as.POSIXlt() *REALLY* is there exactly for this purpose ??? I rather think the authors of each of the other old-fashioned date classes should provide as.POSIXlt() methods for their classes. Then, we'd have uniform interfaces, following's Gabor's # 1. above. My proposal above was just a direction to a common way of dealing with dates within R. If as.POSIXlt() methods is the way, that is perfectly fine with me. -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] mcmc pack
Mariagiulia Matteucci matteucci at stat.unibo.it writes: Hello, I would like to see the C++ code for the MCMC package, how can I do that? Thanks, Hi, - go to CRAN page i.e. http://cran.r-project.org/ or one of its mirros - click on packages - find MCMCpack and click on its link - on MCMCpack page you will notice several links and one of them is *.tar.gz, which is a source for the package - download and unpack that file and you are in the bussines Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] day, month, year functions
Gabor Grothendieck ggrothendieck at gmail.com writes: Here are three ways: xx - as.Date(2006-01-05) # 1. use as.POSIXlt as.POSIXlt(xx)$mday as.POSIXlt(xx)$mon + 1 as.POSIXlt(xx)$year + 1900 # 2. use format as.numeric(format(xx, %d)) as.numeric(format(xx, %m)) as.numeric(format(xx, %Y)) # 3. use month.day.year in chron package library(chron) month.day.year(unclass(xx))$day month.day.year(unclass(xx))$month month.day.year(unclass(xx))$year Hi, it would really be great if there would be sec(), min(), hour() day(), month(), year() generic functions that would work on all date classes. Where applicable of course. I imagine that argument to get out integer or character would alse be nice. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Constrain coefs. in linear model to sum to 0
Hello, [EMAIL PROTECTED] wrote: Gorjanc Gregor asks: Hello! I would like to use constrain to sum coeficients of a factor to 0 instead of classical corner contraint i.e. I would like to fit a model like lm(y ~ 1 + effectA + effectB) and say get parameters intercept effectA_1 effectA_2 effectB_1 effectB_2 effectB_3 where effectA_1 represents deviation of level A_1 from intercept and sum(effectA_1, effectA_2) = 0 and the same for factor B. Is this possible to do? Yes, but not quite as simply as you would like. If you set options(contrasts = c(contr.sum, contr.poly)) for example, then factor models are parametrised as you wish above, BUT you don't get all the effects directly In your case above, for example, if fm is the fitted model object, then coef(fm) Will give you intercept, effectA_2, effectB_2, effectB_3. The remaining effects*_1 you will need to calculate as the negative of the sum of all the others. This gets a bit more complicated when you have crossed terms, a*b, but the same principle applies. Thank you for the response. Peter Dalgaard already mentioned that I can get missing coefs with taking negative of the sum of displayed coefs. However, as you already noted that, things are complicated with crossed terms. I was able to handle nested regression but did not had any luck with interactions. For example: ### --- Picture of the model --- if(FALSE) { | a1 | a2 | -- b1 | 8 | 14 | 11 -- b2 | 13 | 17 | 15 -- b3 | 2 | 8 | 5 -- b4 | 4 | 10 | 7 -- | 6.8 | 12.3 | } ### --- Setup --- N - 400 ab - c( 8, 13, 2, 4, 14, 17, 8, 10) sigmaE - 0.01 levA - c(a1, a2) facA - factor(sample(rep(levA, times=N / length(levA levB - c(b1, b2, b3, b4) facB - factor(sample(rep(levB, times=N / length(levB table(facA, facB) facAB - factor(paste(facA, facB, sep=-)) yAB - ab[as.integer(facAB)] ## Create design matrix and simulate y tmp - model.matrix(~ facAB - 1) y - tmp %*% as.matrix(ab) + rnorm(n=N, mean=0, sd=sigmaE) contrasts(facA) - contr.sum contrasts(facB) - contr.sum ### --- Fit the model --- fit - lm(y ~ facA * facB, data=tmpData) coefs - coef(fit) (int - coefs[1]) (yMean - mean(y)) a - coefs[2] (a - c(a, -a)) tapply(y, list(facA), mean) tapply(y, list(facA), mean) - yMean ## Hmm, why is there such a big difference between mean values and ## coefs from the fit? I am doing something wrong here. b - coefs[3:5] (b - c(b, -sum(b))) tapply(y, list(facB), mean) tapply(y, list(facB), mean) - yMean ## Even more strange (ab - coefs[6:8]) ## [EMAIL PROTECTED] tapply(y, list(facA, facB), mean) tapply(y, list(facA, facB), mean) - yMean ## How can I proceed here? ## Using functions proposed by Martin Maechler: fitAov - aov(y ~ facA * facB) model.tables(fitAov, means, se=TRUE) ## vauuu, this is great model.tables(fitAov, effects, se=TRUE) ## also nice and this fits with my raw mean test What am I doing wrong then above with coefficients from the model? -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] if function and apply
John Kane jrkrideau at yahoo.ca writes: I have a dataset just imported from SPSS. It has any number of 99's as missing data and it looks like the next dataset will have custom missing codes. I have abouat 120 variables and an N of 2000. ... # define function fn - function (x a) { you need comma between x and a i.e. (x, a) ^ if (x==a)return (b) else x } How can you return b, if b was not defined before Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] zero values in LHS and RHS
zubin binabina at bellsouth.net writes: Hello! I have a data set where i need to predict hotel stays in a 3 month period by customer. I have stays for each customer in the 3 month period and the previous 3 month period + other variables. these stays contain an integer that ranges from 0 to 10. classic poisson distribution. my question is around zeros. usually we could take logs on the stays and run an OLS model i think an enhancement would be to fit a poisson GLM model. i am confused about handling of zeros. i will have zero values on the RHS and potentially on the LHS. It seems like the poisson GLM model can handle zeros on the LHS and RHS. i guess my question is fitting a poisson GLM, are zeros allowed? If so, how is this estimated with a log link function? I am not sure about the structure of LHS and RHS matrices, but yes, Poisson model can handle zeroes. If it turns out that there are to many zeroes as expected, you might try ZIP model. Log link is the default for Poisson and it links the expected value of Poisson distribution and not the values itself. Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] negative binomial lmer
Ben Bolker bolker at ufl.edu writes: ... I haven't tried it, but you could also consider using a Poisson-lognormal (rather than neg binomial, which is Poisson-gamma) distribution, which might make this all work rather well in lmer: www.cefe.cnrs.fr/esp/TBElston_Parasitology2001.pdf Actually it is very simple lmer(y ~ effA + (1 | effB), family=quasipoisson) i.e. this fits the following model y_ijk ~ Poisson(\lambda_ijk) log(lambda_ijk) = \mu + effaA_i + effB_ij + e_ijk effB_i ~ Normal(0, \sigma^2_b) e_ijk ~ Normal(0, \sigma^2_e) Gregor __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] test
just a test -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] Is there anywhere recycle()?
Hello! I am writting a function, which should recycle one of its arguments if length of the argument is approprate i.e. something like foo - function(x, a) { n - length(x) if(length(a) n) { # recycle a oldA - a a - vector(length=n) a[1:n] - oldA } ## ... return(a) } foo(c(1, 2), a=c(1, 2)) foo(c(1, 2), a=c(1)) I am now wondering if there is any general/generic functions for such task. Thanks! -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Is there anywhere recycle()?
Hi, Gabor Grothendieck wrote: Try: foo2 - function(x, a) cbind(x,a)[,2] thank you for this. It does work to some extent, but not much better than mine foo. foo2(c(1, 2, 3), a=1) [1] 1 1 1 18:14:08 R foo2(c(1, 2, 3), a=c(1,2,3,4)) [1] 1 2 3 4 Warning message: number of rows of result is not a multiple of vector length (arg 1) in: cbind(1, x, a) 18:14:13 R foo2(c(1, 2, 3), a=c(1,2,3)) [1] 1 2 3 18:14:18 R foo2(c(1, 2, 3), a=c(1,2)) [1] 1 2 1 Warning message: number of rows of result is not a multiple of vector length (arg 2) in: cbind(1, x, a) On 7/23/06, Gregor Gorjanc [EMAIL PROTECTED] wrote: Hello! I am writting a function, which should recycle one of its arguments if length of the argument is approprate i.e. something like foo - function(x, a) { n - length(x) if(length(a) n) { # recycle a oldA - a a - vector(length=n) a[1:n] - oldA } ## ... return(a) } foo(c(1, 2), a=c(1, 2)) foo(c(1, 2), a=c(1)) I am now wondering if there is any general/generic functions for such task. Thanks! -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code. -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
Re: [R] Plot fit of a generic function
Yes, that will do. Thank you Gabor! Gabor Grothendieck wrote: You could plot y vs. fitted(y.lm) where y.lm is the output of lm or plot both y and fitted(y.lm) against x on the same chart. On 7/18/06, Gregor Gorjanc [EMAIL PROTECTED] wrote: Hello! Say I have a function, which creates a design matrix i.e. myFunc - function(x) { ret - cbind(x, x*x, x*x*x) colnames(ret) - 1:ncol(ret) return(ret) } n - 200 x - runif(n=n, min=0, max=100) y - myFunc(x) %*% c(1, 0.2, -0.0002) + rnorm(n=n, sd=100) then I can use this in formulae as here (fit - lm(y ~ myFunc(x))) Now I would like to plot data and fitted function on the plot, but I do not want to access each parameter estimate from object fit i.e. I would like to use something similar to abline for linear regression but in a generic way. Is there anything similar to my case? plot(y=y, x=x) ???plotMyFunc??? Thanks! -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code. -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] Plot fit of a generic function
Hello! Say I have a function, which creates a design matrix i.e. myFunc - function(x) { ret - cbind(x, x*x, x*x*x) colnames(ret) - 1:ncol(ret) return(ret) } n - 200 x - runif(n=n, min=0, max=100) y - myFunc(x) %*% c(1, 0.2, -0.0002) + rnorm(n=n, sd=100) then I can use this in formulae as here (fit - lm(y ~ myFunc(x))) Now I would like to plot data and fitted function on the plot, but I do not want to access each parameter estimate from object fit i.e. I would like to use something similar to abline for linear regression but in a generic way. Is there anything similar to my case? plot(y=y, x=x) ???plotMyFunc??? Thanks! -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 and provide commented, minimal, self-contained, reproducible code.
[R] Efficiency of subsetting with x[someCode] or test-someCode and x[test]
Hello! Does anyone have any knowledge on efficiency of subsetting via x[someCode] or ind - someCode x[ind] In first approach I find it sometime cumbersome if someFunc() is something complex i.e. long code, but it is short otherwise. I wonder if assigning result to some indicator object is less efficient since copies result is first stored and then reused. Is there any general way with this that one should go for? Thanks! -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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] Data in R for tag clouds
Hello! I have written function for tag clouds[1]. Currently it is in alpha state and can be found at[2]. I would like to provide some data for this in examples, but I would like to consider if there is any relevant data for such a plot already in R so I can easily load it. Btw. If anyone has any comments, suggestions, (check also [3]) etc. please feel free to send them to me! Thanks! [1]http://en.wikipedia.org/wiki/Tag_cloud [2]http://www.bfro.uni-lj.si/MR/ggorjan/software/R/index.html#tagCloud [3]http://www.bfro.uni-lj.si/MR/ggorjan/software/R/tagCloud/TODO.txt -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 galleries
Hello! I just noticed new link on R wiki on R galleries and wanted to share this info with YOU! - R graphical manuals (this is awesome page as there are all help pages of all packages on CRAN and probably even more and all graphics examples are displayed! - more than 8000 images!) http://bg9.imslab.co.jp/Rhelp/ This is a very nice addition to already existing R graph and movies galleries - R Graph Gallery http://addictedtor.free.fr/graphiques/ - R Movies Gallery http://addictedtor.free.fr/movies/ -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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] Full or non constrained/reparametrized model.matrix
Gabor thanks! Last suggestion works like charm! Gabor Grothendieck wrote: On 5/9/06, Gabor Grothendieck [EMAIL PROTECTED] wrote: On 5/9/06, Gregor Gorjanc [EMAIL PROTECTED] wrote: Hello! Thank you very much for the response. Please read within the lines Gabor Grothendieck wrote: On 5/9/06, Gregor Gorjanc [EMAIL PROTECTED] wrote: Hello! I have parameter estimates for a generalized linear model and would like to produce fitted values i.e. fitted(). This can be easily done in R, but my problem lies in fact that I have a vector of parameters from some other software, that is is not constrained i.e. I have the following estimates for model with one factor with 4 levels beta = c(intercept group1 group2 group3 group4) where group1:4 are estimated deviations from intercept i.e. sum to zero contraint, but all parameter estimates are there! How can I create a model matrix that will not contain any constraints since I would like to compute model.matrix(some_formula) %*% beta I.e. I would like to have model.matrix of the form 1 1 0 0 0 1 0 1 0 0 1 0 0 1 0 1 0 0 0 1 and not of the following form with contr.treatment or any other contraints 1 0 0 0 1 1 0 0 1 0 1 0 1 0 0 1 I could remove group4 from beta and use sum to zero contraint for contrast in fomula, but I would like to overcome this, as my model can be richer in number or parameters. The following example, will show what I would like to do: ## --- Setup --- groupN - 4 NPerGroup - 5 min - 1 max - 5 g - runif(n = groupN, min = min, max = max) ## --- Simulate --- group - factor(rep(paste(G, 1:groupN, sep = ), each = NPerGroup)) y - vector(mode = numeric, length = groupN * NPerGroup) j - 1 for (i in 1:groupN) { y[j:(i * NPerGroup)] - rpois(n = NPerGroup, lambda = g[i]) j - (i * NPerGroup) + 1 } ## --- GLM --- contrasts(group) - contr.sum(groupN) fit - glm(y ~ group, family = poisson) coef(fit) ## Now this goes nicely model.matrix(y ~ group) %*% coef(fit) ## But pretend I have the following vector of estimated parameters beta - c(coef(fit), 0 - sum(coef(fit)[-1])) names(beta) - c(names(beta)[1:4], group4) ## I can not apply this as model matrix does not conform to beta model.matrix(y ~ group) %*% beta Try this: model.matrix(y ~ group-1) %*% beta[-1] + beta[1] This is a nice hack here, but does not work in general. Say I have another factor sex - factor(rep(paste(S, 1:2, sep = ), times=10)) model.matrix(y ~ group + sex - 1) produces a matrix of 5 columns and not 6 as I want to. Try this: cbind(1, model.matrix(~group-1), model.matrix(~sex-1)) In thinking about this a bit more, try this: attr(group, contrasts) - diag(nlevels(group)) attr(sex, contrasts) - diag(nlevels(sex)) model.matrix(~ group + sex) -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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] Full or non constrained/reparametrized model.matrix
Hello! I have parameter estimates for a generalized linear model and would like to produce fitted values i.e. fitted(). This can be easily done in R, but my problem lies in fact that I have a vector of parameters from some other software, that is is not constrained i.e. I have the following estimates for model with one factor with 4 levels beta = c(intercept group1 group2 group3 group4) where group1:4 are estimated deviations from intercept i.e. sum to zero contraint, but all parameter estimates are there! How can I create a model matrix that will not contain any constraints since I would like to compute model.matrix(some_formula) %*% beta I.e. I would like to have model.matrix of the form 1 1 0 0 0 1 0 1 0 0 1 0 0 1 0 1 0 0 0 1 and not of the following form with contr.treatment or any other contraints 1 0 0 0 1 1 0 0 1 0 1 0 1 0 0 1 I could remove group4 from beta and use sum to zero contraint for contrast in fomula, but I would like to overcome this, as my model can be richer in number or parameters. The following example, will show what I would like to do: ## --- Setup --- groupN - 4 NPerGroup - 5 min - 1 max - 5 g - runif(n = groupN, min = min, max = max) ## --- Simulate --- group - factor(rep(paste(G, 1:groupN, sep = ), each = NPerGroup)) y - vector(mode = numeric, length = groupN * NPerGroup) j - 1 for (i in 1:groupN) { y[j:(i * NPerGroup)] - rpois(n = NPerGroup, lambda = g[i]) j - (i * NPerGroup) + 1 } ## --- GLM --- contrasts(group) - contr.sum(groupN) fit - glm(y ~ group, family = poisson) coef(fit) ## Now this goes nicely model.matrix(y ~ group) %*% coef(fit) ## But pretend I have the following vector of estimated parameters beta - c(coef(fit), 0 - sum(coef(fit)[-1])) names(beta) - c(names(beta)[1:4], group4) ## I can not apply this as model matrix does not conform to beta model.matrix(y ~ group) %*% beta ## Is there any general way of constructing full design/model matrix ## without any constraints/reparametrizations? Thanks! -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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] Full or non constrained/reparametrized model.matrix
Hello! Thank you very much for the response. Please read within the lines Gabor Grothendieck wrote: On 5/9/06, Gregor Gorjanc [EMAIL PROTECTED] wrote: Hello! I have parameter estimates for a generalized linear model and would like to produce fitted values i.e. fitted(). This can be easily done in R, but my problem lies in fact that I have a vector of parameters from some other software, that is is not constrained i.e. I have the following estimates for model with one factor with 4 levels beta = c(intercept group1 group2 group3 group4) where group1:4 are estimated deviations from intercept i.e. sum to zero contraint, but all parameter estimates are there! How can I create a model matrix that will not contain any constraints since I would like to compute model.matrix(some_formula) %*% beta I.e. I would like to have model.matrix of the form 1 1 0 0 0 1 0 1 0 0 1 0 0 1 0 1 0 0 0 1 and not of the following form with contr.treatment or any other contraints 1 0 0 0 1 1 0 0 1 0 1 0 1 0 0 1 I could remove group4 from beta and use sum to zero contraint for contrast in fomula, but I would like to overcome this, as my model can be richer in number or parameters. The following example, will show what I would like to do: ## --- Setup --- groupN - 4 NPerGroup - 5 min - 1 max - 5 g - runif(n = groupN, min = min, max = max) ## --- Simulate --- group - factor(rep(paste(G, 1:groupN, sep = ), each = NPerGroup)) y - vector(mode = numeric, length = groupN * NPerGroup) j - 1 for (i in 1:groupN) { y[j:(i * NPerGroup)] - rpois(n = NPerGroup, lambda = g[i]) j - (i * NPerGroup) + 1 } ## --- GLM --- contrasts(group) - contr.sum(groupN) fit - glm(y ~ group, family = poisson) coef(fit) ## Now this goes nicely model.matrix(y ~ group) %*% coef(fit) ## But pretend I have the following vector of estimated parameters beta - c(coef(fit), 0 - sum(coef(fit)[-1])) names(beta) - c(names(beta)[1:4], group4) ## I can not apply this as model matrix does not conform to beta model.matrix(y ~ group) %*% beta Try this: model.matrix(y ~ group-1) %*% beta[-1] + beta[1] This is a nice hack here, but does not work in general. Say I have another factor sex - factor(rep(paste(S, 1:2, sep = ), times=10)) model.matrix(y ~ group + sex - 1) produces a matrix of 5 columns and not 6 as I want to. ## Is there any general way of constructing full design/model matrix ## without any constraints/reparametrizations? Thanks! -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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] Making R talk to Win/OpenBUGS in Linux (again)
Hello Paul, Wow! That was quick. One more thing would make my R2WinBUGS experience more perfect. I wish the bugs function explained how the user can write out the model.txt file from within the R code itself, rather than taking the current advice of creating it separately. The process now escribed in bugs is: 1. Write a WinBUGS model in a ASCII file. 2. Go into R. I don't know what the best way to create model.txt might be, but this does work, and maybe you have even better examples to show: myTextExample - c( model { for (j in 1:J){ y[j] ~ dnorm (theta[j], tau.y[j]) theta[j] ~ dnorm (mu.theta, tau.theta) tau.y[j] - pow(sigma.y[j], -2) } mu.theta ~ dnorm (0.0, 1.0E-6) tau.theta - pow(sigma.theta, -2) sigma.theta ~ dunif (0, 1000) } ) zz - file(model.txt,w) writeLines(myTextExample,con=zz) close(zz) The only dicey part here is that the user might save the model in the wrong directory, but I don't see what prevents that with the current advice in bugs I suggest this because it allows the creation of one file working examples in R2WinBUGS. For the teaching objective, this simplifies things a lot! Perhaps you might try my function, which does this exactly as you proposed. I added also a print function so you can print model nicely also in R. You can find documentation here http://www.bfro.uni-lj.si/MR/ggorjan/software/R/ggmisc/ggmiscHtml/bugsModel.html and the function itself lives in my offCRAN package ggmisc, which you can grab at http://www.bfro.uni-lj.si/MR/ggorjan/software/R/ggmisc.tar.gz I already proposed this to Uwe, but he did not find it enough interesting to include it into the R2WinBUGS. Anyway, Jouni Kerman, has written even better function - fitbugs(). Look at link bellow for more. I hope fitbugs() will make it into one of next releases of R2WinBUGS. http://www.stat.columbia.edu/~kerman/Software/index.html On 5/4/06, Uwe Ligges [EMAIL PROTECTED] wrote: Paul Johnson wrote: Hello, and thanks for the many very rapid responses to my requests for R/WinBUGS conversation. I do not know how to build OpenBUGS in linux, but will snoop around and see if I can't figure out what's wrong with linking R to BRugs.so. In the meanwhile... While we are on the linux/bugs topic, can I please ask for a fix in R2WinBUGS? On my system, the script.txt file is written into the WinBUGS directory,where the user does not ordinarily have write permissions. If script.txt could be written with the other temporary files, this would be solved. Will be fixed for the next release. We should also add some option to disable running bugs.update.settings() which really requires write access in the WinBUGS directory. -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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] New package connectedness
Dear R users, I would like to announce a new package connectedness on CRAN. connectedness can be used to identify disconnected sets. It contains functions to find, plot and subset disconnected sets in a two-way classification without interaction. Please do not hesitate to contact me, if you find a bug (possibly with a patch) or you have any comments. -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. ___ 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] How to specify function arguments that are used in different places
Hello! Subject is not very clear, but I hope my question will be;) I wrote a function, which produces a plot and I have problems with arguments. For the sake of example let us consider that my function looks like this myfunc - function(x, points=FALSE, lines=FALSE, ...) { ## x is an object that is being plotted plot(x$plotData, ...) ## one can also add some data on graph via points points(x$pointsData, ...) ## one can also add some data on graph via lines lines(x$linesData, ...) } My problem is in ... argument. plot(), points() and lines() have so many possible arguments, which is very nice, but how can I deal with them in my scenario. For example, I might want to specify red color for plot, blue for points and green for lines. Is it possible to handle such a mixture, without specifiying zillion of arguments such as plotCol, pointsCol, linesCol etc.? Perhaps something like ~ points$...? Thanks! -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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] Making R talk to Win/OpenBUGS in Linux (again)
Hello Jun, On 5/1/06, jun yan [EMAIL PROTECTED] wrote: I have used linbugs with the rbugs package for a recent work. It might be worthwhile trying. It would be very nice, if you would consider to port new funcionality in rbugs to R2WinBUGS. I thought that this was your intention, when wine part was added to R2WinBUGS. I do appreciate your package, but find it weird to have two packages i.e. R2WinBUGS and rbugs for exactly the same task. The issue with all this packages is really getting very comples, now that BRugs is also availabe. Please take this only as a humble suggestion. Looking forward to hear from you. -- Lep pozdrav / With regards, Gregor Gorjanc University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzaletel: +386 (0)1 72 17 861 Slovenia, Europe fax: +386 (0)1 72 17 888 One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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] Making R talk to Win/OpenBUGS in Linux (again)
= 1000,bugs.directory = /usr/local/share/WinBUGS14/, working.directory = /tmp, clearWD = FALSE, useWINE=T, newWINE=T, WINE=/usr/bin/wine,WINEPATH=/usr/bin/winepath) I do have the binary /usr/bin/winepath, but can't tell how to give R2WinBUGS what it wants. Many failed efforts below: schools.sim - bugs(data, inits, parameters, model.file, n.chains = 3, n.iter = 1000,bugs.directory = /usr/local/share/WinBUGS14/, working.directory = /tmp, clearWD = FALSE, useWINE=T, newWINE=T, WINE=/usr/bin/wine,WINEPATH=/usr/bin/) Error in paste(WINEPATH, -w, x) : object WINEPATH not found schools.sim - bugs(data, inits, parameters, model.file, n.chains = 3, n.iter = 1000,bugs.directory = /usr/local/share/WinBUGS14/, working.directory = /tmp, clearWD = FALSE, useWINE=T, newWINE=T, WINE=/usr/bin/wine) Error in if (!nchar(WINEPATH)) { : argument is of length zero schools.sim - bugs(data, inits, parameters, model.file, n.chains = 3, n.iter = 1000,bugs.directory = /usr/local/share/WinBUGS14/, working.directory = /tmp, clearWD = FALSE, useWINE=T, newWINE=T, WINE=/usr/bin/wine,WINEPATH=/usr/bin) Error in paste(WINEPATH, -w, x) : object WINEPATH not found I know there are many unknowns here, but hope somebody has at least one working example that we can build on. -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. -- __ 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] Remove [1] ... from output
Hello! I am writing some numbers and character vectors to an ascii file and would like to get rid of [1] ... as shown bellow (a dummy example) R runif(20) [1] 0.653574 0.164053 0.036031 0.127208 0.134274 0.103252 0.506480 0.547759 [9] 0.912421 0.584382 0.987208 0.996846 0.666760 0.053637 0.327590 0.370737 [17] 0.505706 0.412316 0.887421 0.812151 I have managed to work up to remove quotes and all [*] except [1] as shown bellow. R print(paste(runif(20), collapse = ), quote = FALSE) [1] 0.790620362851769 0.45603066496551 0.563822037540376 0.812907998682931 0.726162418723106 0.37031230609864 0.681147597497329 0.29929908295162 0.209858040558174 0.304300333140418 0.105796672869474 0.743657597573474 0.409294542623684 0.825012607965618 0.282235795632005 0.21159387845546 0.620056127430871 0.337449935730547 0.754527133889496 0.280175548279658 Any hints how to solve my task? -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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] Remove [1] ... from output
Thank you Adelchi and Ales! Adelchi Azzalini wrote: On Tue, 28 Mar 2006 13:21:59 +0200, Gregor Gorjanc wrote: GG Hello! GG GG I am writing some numbers and character vectors to an ascii file GG and would like to get rid of [1] ... as shown bellow (a dummy GG example) GG GG R runif(20) GG [1] 0.653574 0.164053 0.036031 0.127208 0.134274 0.103252 GG 0.506480 0.547759 [9] 0.912421 0.584382 0.987208 0.996846 GG 0.666760 0.053637 0.327590 0.370737 GG [17] 0.505706 0.412316 0.887421 0.812151 GG GG I have managed to work up to remove quotes and all [*] except [1] GG as shown bellow. GG GG R print(paste(runif(20), collapse = ), quote = FALSE) GG [1] 0.790620362851769 0.45603066496551 0.563822037540376 GG 0.812907998682931 0.726162418723106 0.37031230609864 GG 0.681147597497329 0.29929908295162 0.209858040558174 GG 0.304300333140418 0.105796672869474 0.743657597573474 GG 0.409294542623684 0.825012607965618 0.282235795632005 GG 0.21159387845546 0.620056127430871 0.337449935730547 GG 0.754527133889496 0.280175548279658 GG GG Any hints how to solve my task? GG what about cat(format(runif(20))) ? you can improve it a bit with cat(format(runif(20)),\n) -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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] Mixed-Effects Models in S and S-Plus (Springer)
Hello! Book by Pinheiro and Bates is being repeatedly suggested in connection with lme() from nlme package. Is this book also suggested for a reference use with lmer in lme4 package? -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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] Mixed-Effects Models in S and S-Plus (Springer)
Doran, Harold wrote: Well, not if you are interested in lmer() syntax. The book is good with the conceptual issues in modeling the data, but does not include any reference to fitting models with lmer. OK, I of course suspected that lmer will not be in, but I wanted to hear that the book has good conceptual issues. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Gregor Gorjanc Sent: Monday, March 20, 2006 8:58 AM To: [EMAIL PROTECTED] Subject: [R] Mixed-Effects Models in S and S-Plus (Springer) Hello! Book by Pinheiro and Bates is being repeatedly suggested in connection with lme() from nlme package. Is this book also suggested for a reference use with lmer in lme4 package? -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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 -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ 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