[R] Plot Point Labelling
Hi, Please tell me of an R graphical routine that will create a 3d plot something like cloud or scatterplot3d but that allows individual labeling of each point in the plot. I've done it with SPSS. I'm sure that there is an R routine, but I can't find it. I'm at: [EMAIL PROTECTED] [EMAIL PROTECTED] Thanks! Charles Charles Liard CMHA Vancouver - Burnaby Branch Tel: (604) 872 4902 This email is intended only for the person to whom it was addressed (the addressee) and may contain confidential and / or private material. Any retransmission, dissemination, or other communication to a person other than the addressee of this message is prohibited. If you received this email in error, please contact the sender (or call 604 872 4902) and destroy all copies of the message. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] A question about pairs()
Greetings R users, I am an R graphics newbie trying to produce a custom trellis plot using pairs() with R 2.7.2. I have spatial data on which I run a geographically weighted regression (gwr, using the -spgwr- package). I want to check the gwr coefficients for multicollinearity and spatial association, following Wheeler and Tiefelsdorf (2005), and I would like to summarize the results of this exercise using -pairs-. My problem is that I want to generate figures for two groups of variables and then assemble them into a panel display. The first set of variables is the gwr coefficients, for which I want to generate x-y scatterplots. This is straightforward, and works fine. The second set of variables is the local correlation between the gwr coefficient estimates, which I want to plot as kernel densities. Is there any way to do this? I can do one or the other, but not both together. For example: data(columbus) # compute gwr bandwidth col.bw - gwr.sel(crime ~ income + housing, data=columbus, coords=cbind(columbus$x, columbus$y)) # estimate gwr col.gauss - gwr(crime ~ income + housing, data=columbus, coords=cbind(columbus$x, columbus$y), bandwidth=col.bw, hatmatrix=TRUE) # summarize gwr results col.gauss # compute local correlation matrix among gwr coefs col.cor - gw.cov(col.gauss$SDF,vars=2:4,bw=col.bw,cor=T,longlat=F) # extract gwr coefficients and local correlations col.gauss.mat - as(col.gauss$SDF, data.frame)[,2:4] col.cor.mat - as(col.cor$SDF, data.frame)[,16:18] # define correlation function for use by -pairs- panel.cor - function(x,y,digits=4) { r = (cor(x,y,use=pairwise)) txt - format(c(r, 0.123456789), digits=digits)[1] text(0.5, 0.5, txt) } # show gwr results: works fine, except that text positioning is screwed up pairs(col.gauss.mat, upper.panel=panel.smooth, lower.panel=panel.cor, labels=c(Intercept,Income,Housing)) # show gwr local correlations: works fine, except that text positioning is screwed up pairs(col.cor.mat, upper.panel=panel.smooth, lower.panel=panel.cor, labels=c(Intercept,Income,Housing)) # My problem is with the following function... panel.dis - function(...) { if(panel.number()==4) dist - lines(density(col.cor.mat[,1])) if(panel.number()==7) dist - lines(density(col.cor.mat[,2])) if(panel.number()==8) dist - lines(density(col.cor.mat[,3])) } pairs(col.gauss.mat, upper.panel=panel.smooth, lower.panel=panel.dis, labels=c(Intercept,Income,Housing)) The last -pairs- call terminates with the following error: Error in if (panel.number() == 4) dist - lines(density(col.cor.mat[, : argument is of length zero I thought I could address panels in this way using the panel.number() function. I would very much appreciate it if someone could let me know what I am doing wrong. Should I not use -pairs- at all? The problem is that I'd like the plots to line up nicely. -i -- Ian Sue Wing 675 Commonwealth Ave., Boston MA 02215 Associate Professor Tel: (617) 353-5741 Dept. of Geography Environment Fax: (617) 353-5986 Boston University Web: http://people.bu.edu/isw __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] PAM: how to get the best number of clusters
I have the book you mentioned. It basically describes the silhouette method. I do not have it handy as I moved so it is still in some box. However I cannot remember that book providing any other criterion to find the best clusters number. On the other hand I have the same problem with hierarchical clustering techniques. I use clusters as exploratory analysis because I do not have any a-priori knowledge that helps me make a choice. How can multivariate analysis help? I launched a loop where the silhouette test follows PAM which is passed a clusters number increased by 1 at each iteration. Since I am observing that the silhouette value is now oscillating among negative numbers, I wonder whether I can assume that it can only grow worse once it has turned negative the first time so leave the loop after the first negative number and choose the clusters number associated with the biggest positive silhouette value. This procedure would spare a lot of CPU time. Thank you very much, Maura On Thu, Oct 30, 2008 at 7:25 PM, Dylan Beaudette [EMAIL PROTECTED]wrote: On Thursday 30 October 2008, Maura E Monville wrote: I have a pretty big similarity matrix (2870x2870). I will produce even bigger ones soon. I am using PAM to generate clusters. The desired number of output clusters is a PAM input parameter. I do not know a-priopri what is the best clusters layout . I resorted to the silhouette test. It takes forever as I have to run PAM with all possible numbers of clusters. I wonder whether there is some faster method, either a s/w code or some theoretical guidelines, to get the optimum clusters number. Thank you very much, This is a very general topic in the field of multivariate analysis. There really isn't any way to know the 'correct' number of clusters, however there are several metrics that can give you an indication of how messy your data are. For information on the methods in the cluster package, see this book: Kaufman, L. Rousseeuw, P. J. Finding Groups in Data An Introduction to Cluster Analysis Wiley-Interscience, 2005 Otherwise, consider a book on multivariate analysis. Alternatively, try a hierarchical clustering approach, and look for meaningful groupings. Some thing like this: d - diana(daisy(your_data_matrix)) d.hc - as.hclust(d) d.hc$labels - your_data_matrix$id plot(d.hc) Cheers, Dylan -- Dylan Beaudette Soil Resource Laboratory http://casoilresource.lawr.ucdavis.edu/ University of California at Davis 530.754.7341 -- Maura E.M [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Problems with R make - related to blas
Hello, I am having problems installing R-2.7.2 on a maching running OpenSUSE Linux 10.2. Initially, R couldn't find a Fortran compiler, so I downloaded and installed compat-g77-3.3.5-38.i586.rpm. I then set FPICFLAGS=fPIC in the config.site file. I next configured using ./configure, with no errors. When I run make, I get the following error: make[3]: Entering directory `/usr/local/src/R-2.7.2/src/extra/blas' make[4]: Entering directory `/usr/local/src/R-2.7.2/src/extra/blas' g77 fPIC -g -c blas.f -o blas.o g77: cannot specify -o with -c or -S and multiple compilations make[4]: *** [blas.o] Error 1 make[4]: Leaving directory `/usr/local/src/R-2.7.2/src/extra/blas' make[3]: *** [R] Error 2 make[3]: Leaving directory `/usr/local/src/R-2.7.2/src/extra/blas' make[2]: *** [R] Error 1 make[2]: Leaving directory `/usr/local/src/R-2.7.2/src/extra' make[1]: *** [R] Error 1 make[1]: Leaving directory `/usr/local/src/R-2.7.2/src' make: *** [R] Error 1 I'd appreciate it if anyone could guide me around this problem... Thanks, William __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] R help for invoking nmmin()
My code is as follows: #include iostream #include cmath using namespace std; #define MATHLIB_STANDALONE 1 extern C { #include R_ext/Applic.h } typedef struct TT{ double ** tempX; double * tempY; int tempN; } TT, *MM; double fn(int N, double * beta, void * ex){ double total = 0; int i = 0,j = 0; double * betaFn = new double[N]; MM tmp = (MM)ex; for(i = 0; i N; i++) { betaFn[i] = exp(beta[i]); } for(j = 0; j tmp-tempN; j++) { double temp = 0; for(i = 0; i N; i ++) { temp += tmp-tempX[j][i] * betaFn[i]; } total += pow(tmp-tempY[j] - temp,2); } return total; } void mgr(int, double *, double *, void *){} int main() { //coutpchisq(2,7,1,0)endl; //cout qnorm(0.7, 0.0, 1.0, 0, 0) endl; int n = 5; double *dpar, *opar; int fncount, fail; dpar = new double [n]; opar = new double [n]; dpar[0] = 13.67318; dpar[1] = 0; dpar[2] = 17.02707; dpar[3] = 0; dpar[4] = 24.08231; double value; double abstol = 1e-16; double intol = 1e-8; int y = 12; MM t = (MM)malloc(y*(n+2)*sizeof(TT)); t- tempX = new double * [y]; for(size_t ii = 0; ii y; ii++) { t-tempX[ii] = new double [n]; } t-tempX[0][1] = -0.03; t-tempX[1][1] = -0.02; t-tempX[2][1] = -0.015; t-tempX[3][1] = -0.01; t-tempX[4][1] = -0.005; t-tempX[5][1] = 0; t-tempX[6][1] = 0.005; t-tempX[7][1] = 0.01; t-tempX[8][1] = 0.015; t-tempX[9][1] = 0.02; t-tempX[10][1] = 0.03; t-tempX[11][1] = 0.04; for(int ii = 0; ii n; ii++) { for(int yy = 0; yy y; yy++) t-tempX[yy][ii] = pow(t-tempX[yy][1], ii); } t- tempY = new double [y]; t-tempY[0] = 930862; t-tempY[1] = 893736; t-tempY[2] = 882721; t-tempY[3] = 872066; t-tempY[4] = 871729; t-tempY[5] = 868550; t-tempY[6] = 865100; t-tempY[7] = 862132; t-tempY[8] = 870266; t-tempY[9] = 870493; t-tempY[10] = 893980; t-tempY[11] = 952914; t-tempN = y; cout fn(n,dpar,(void*)t)endl; nmmin(n, dpar, opar, value, fn, fail, abstol, intol, (void *)t, 1, 0.5, 2, 0, fncount, 500); for(int aa = 0 ; aa n; aa++) { cout opar[aa],; } cout\n; return 0; } I run this code, It told me Segmentation fault. I dont know if the parameter is wrong or not. Thank you very much. DJade -- View this message in context: http://www.nabble.com/R-help-for-invoking-nmmin%28%29-tp20258786p20258786.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Problems installing R-2.8.0 on Suse linux 10.1
Hello, I am trying to install R on Suse linux 10.1. To get configure to run correctly, I had to set FPICFLAGS=fPIC. R is now configured for i686-pc-linux-gnu Source directory: . Installation directory:/usr/local C compiler:gcc -std=gnu99 -g -O2 Fortran 77 compiler: g77 -g C++ compiler: g++ -g -O2 Fortran 90/95 compiler:g77 -g Obj-C compiler: Interfaces supported: X11 External libraries:readline Additional capabilities: PNG, JPEG, TIFF, iconv, MBCS, NLS Options enabled: shared BLAS, R profiling, Java Recommended packages: yes When I then run make, I get: make[3]: Entering directory `/usr/local/src/R-2.8.0/src/extra/blas' make[4]: Entering directory `/usr/local/src/R-2.8.0/src/extra/blas' g77 fPIC -g -c blas.f -o blas.o g77: cannot specify -o with -c or -S and multiple compilations make[4]: *** [blas.o] Error 1 make[4]: Leaving directory `/usr/local/src/R-2.8.0/src/extra/blas' make[3]: *** [R] Error 2 make[3]: Leaving directory `/usr/local/src/R-2.8.0/src/extra/blas' make[2]: *** [R] Error 1 make[2]: Leaving directory `/usr/local/src/R-2.8.0/src/extra' make[1]: *** [R] Error 1 make[1]: Leaving directory `/usr/local/src/R-2.8.0/src' make: *** [R] Error 1 Help would be appreciated, William __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Plot Point Labelling
Check out panel.identify.cloud in the latest version of the lattice package. E.g. cloud(depth ~ lat * long, quakes) trellis.focus() panel.identify.cloud() 2008/10/31 Charles Liard [EMAIL PROTECTED]: Hi, Please tell me of an R graphical routine that will create a 3d plot something like cloud or scatterplot3d but that allows individual labeling of each point in the plot. I've done it with SPSS. I'm sure that there is an R routine, but I can't find it. I'm at: [EMAIL PROTECTED] [EMAIL PROTECTED] Thanks! Charles Charles Liard CMHA Vancouver - Burnaby Branch Tel: (604) 872 4902 This email is intended only for the person to whom it was addressed (the addressee) and may contain confidential and / or private material. Any retransmission, dissemination, or other communication to a person other than the addressee of this message is prohibited. If you received this email in error, please contact the sender (or call 604 872 4902) and destroy all copies of the message. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Felix Andrews / 安福立 http://www.neurofractal.org/felix/ 3358 543D AAC6 22C2 D336 80D9 360B 72DD 3E4C F5D8 __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Convert windows package to Linux source package
Indeed in the case of a lazy-loaded package it looks a bit tricky, but it should still be possible, as I guess one can access the source code of any function defined in a package, once loaded (am I wrong?). When no lazy-load takes place then it is actually quite straight forward as you said Duncan: - I looked into the merged Rd file, and the functions documentation appear embedded between as follows: % --- Source file: function-name.Rd ... ... \eof So with some shell scripting (csplit, sed, ...) I succeeded in separate the single file into separate Rd files, named after the functions' names. - the source files can be generated using package.skeleton on the R source files (remember no lazy-load, so plain and complete source code is directly available) - then replace the generic Rd files generated by package.skeleton by the Rd files obtained from the merged Rd file. - copy the original DESCRIPTION file into the source package directory, removing the 'Build' section. In theory this works well, except if: - the window packages Rd file contains error (!): I couldn't believe it but some of the Rd definitions where wrong (missing closing bracket, invalid tags (reference instead of references), invalid keywords, ...). Isn't it mandatory to have valid Rd files to be able to build a package with R CMD build? - the package's dependencies are not defined in the original DESCRIPTION file. Isn't it also mandatory to specify if the package depends on other packages ? (because R CMD check can effectively detect the discrepancy between the field DESCRIPTION:Dependencies and what is directly required by the package...) Prof Brian Ripley wrote: On Thu, 30 Oct 2008, Duncan Murdoch wrote: On 10/30/2008 10:50 AM, Renaud Gaujoux wrote: Hi, is there a simple way, I mean a utility that converts a windows package (.zip file) to a Linux source package (tar.gz or even simply extract the source code and Rd files needed to run successfully a R CMD check). The windows package doesn't have any external lib, C code, etc..., just plain R code, so in theory it is quite straight forward (extract the code from the functions, and re-separate the Rd files from the man directory. Any idea? No, no such utility exists as far as I know. But as you say, it should be relatively straightforward to write one: the Rd files are concatenated and compressed into man/pkg.Rd.gz, and the R files are concatenated into R/pkg. Not if lazy-loading is used (and it normally is). But what really is wrong with Roger Koenker's suggestion: 'unzip -a' the .zip into a library directory? It will most likely work (the reverse does: zip up on Linux, unzip on Windows). You just need to watch out for comments in the files that will be mistaken for separators. I think that is safe enough in Rd files, as we process the concatenated versions. Duncan Murdoch __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Problems installing R-2.8.0 on Suse linux 10.1
On Thu, 30 Oct 2008, William Hudspeth wrote: Hello, I am trying to install R on Suse linux 10.1. To get configure to run correctly, I had to set FPICFLAGS=fPIC. FPICFLAGS=-fPIC would have been correct (or -fpic). It is odd that FFLAGS has been set to -g and not '-g -O2': did you change that? And also odd that FC has been set to g77, which is not a Fortran 90/95 compiler. Your problem was a direct result of your incorrect setting. However, I am rather surprised that any change was needed, as R's configure correctly finds FPICFLAGS on many Linux systems, and I am sure has in the past on Suse 10.1. R is now configured for i686-pc-linux-gnu Source directory: . Installation directory:/usr/local C compiler:gcc -std=gnu99 -g -O2 Fortran 77 compiler: g77 -g C++ compiler: g++ -g -O2 Fortran 90/95 compiler:g77 -g Obj-C compiler: Interfaces supported: X11 External libraries:readline Additional capabilities: PNG, JPEG, TIFF, iconv, MBCS, NLS Options enabled: shared BLAS, R profiling, Java Recommended packages: yes When I then run make, I get: make[3]: Entering directory `/usr/local/src/R-2.8.0/src/extra/blas' make[4]: Entering directory `/usr/local/src/R-2.8.0/src/extra/blas' g77 fPIC -g -c blas.f -o blas.o g77: cannot specify -o with -c or -S and multiple compilations make[4]: *** [blas.o] Error 1 make[4]: Leaving directory `/usr/local/src/R-2.8.0/src/extra/blas' make[3]: *** [R] Error 2 make[3]: Leaving directory `/usr/local/src/R-2.8.0/src/extra/blas' make[2]: *** [R] Error 1 make[2]: Leaving directory `/usr/local/src/R-2.8.0/src/extra' make[1]: *** [R] Error 1 make[1]: Leaving directory `/usr/local/src/R-2.8.0/src' make: *** [R] Error 1 Help would be appreciated, William __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Suppressing internal grid in filled.contour
Martin Maechler maechler at stat.math.ethz.ch writes: Yes, indeed, PDFs *are* fine Apologies to the original poster, I was not aware of this. Dieter __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Problem of running R console from Java on linux box
Hi the list: I can not create R process from Java application with my linux box, the process never was created, or just terminated immediately after I called Runtime.getRuntime().exec(...) The command lines I have tried are(the paths are all correct): /bin/sh /usr/bin/R /bin/sh -c /usr/bin/R /usr/lib/R/bin/exec/R(also set R_HOME=/usr/lib/R) I also tried ProcessBuilder, which is no help. I'm confused, is this possible? Many Thanks! __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] package pls
tsn4867 [EMAIL PROTECTED] writes: For the package pls, I need to understand the algorithm for simpls.fit for Partial Least Squares. I'm not sure if simpls.fit tries to find the weight vectors (loadings) to maximize which of the two: Cov(Xw, y) or maximize Cov^2(Xw,y)? Are these objective functions equivalent? (in some texts, they use the first and in other texts, they use the second obj. function.). I think the algorithm for simpls.fit is using Cov(Xw,y). Also, can you give me some references where they state the equivalency of the two obj. functions? The implementation in simpls.fit follows the algorithm in de Jong, S. (1993) SIMPLS: an alternative approach to partial least squares regression. _Chemometrics and Intelligent Laboratory Systems_, *18*, 251-263. (up to simplifications and performance changes). I don't recall if the criterion was cov or cov^2, but I believe they should be identical (up to sign). -- Bjørn-Helge Mevik __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] how to retrieve some records from a dataframe
Hi , Dataframe Df contains more than 30 no of records.but I want only first 30 rows of data. How can I retrieve it. Thanks K.Ravichandra [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] FW: Re: linux batch question
ML == Mark Leeds [EMAIL PROTECTED] on Wed, 29 Oct 2008 17:13:04 -0500 (CDT) writes: ML Hi Phil: That's EXACTLY what it is. Thanks so much. It's nice to know ML that the R Gods don't hate me. I hope it's okay that I'm going to cc ML r-help ML in case this thread comes up in the future and also so that other people ML who might want to help know that it's solved. Thanks again. another note below, just for the archives i.e. future reference: ML On Wed, Oct 29, 2008 at 5:01 PM, Phil Spector wrote: Mark: delete workspace? (y/n): print(test of batch\n) It looks like you've got something in your .Rprofile or .First that's causing R to prompt you about saving your workspace, and it's interpreting your program as that response. Try R --no-save --vanilla CMD BATCH test.r test.out A mini-remark: I'd recommend *dropping* the final 'test.out', such that R will automatically use test.Rout and the *.Rout ending is automagically recognized by an ESS-enabled emacs [as R Transcript file] and maybe similarly by other R-aware editors. Martin - Phil On Wed, 29 Oct 2008, [EMAIL PROTECTED] wrote: I usually just run my R programs at the R command prompt but for my latest one I want to save any output that gets written to the screen so I am trying to use R CMD BATCH and send the output to an output file. I realize I could use sink at the prompt but I'd rather try to do it this way because I know that I used to do this in S+. So, I wrote a simple one line program called test.R which is below. print(test of batch\n) Then I did R CMD BATCH temp.R temp.out temp.out which is shown below then ends up containing all my loading up messages and a proc.time statement at the very bottom but not the print statement itself ? Does someone know what I'm doing wrong. I've looked around but I can't find anything that answers my question. My sessionInfo() is at ther very bottom of this email. Thanks. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] how to retrieve some records from a dataframe
Df[1:30,] --- On Fri, 31/10/08, Kurapati, Ravichandra (Ravichandra) [EMAIL PROTECTED] wrote: From: Kurapati, Ravichandra (Ravichandra) [EMAIL PROTECTED] Subject: [R] how to retrieve some records from a dataframe To: r-help@r-project.org Date: Friday, 31 October, 2008, 2:14 PM Hi , Dataframe Df contains more than 30 no of records.but I want only first 30 rows of data. How can I retrieve it. Thanks K.Ravichandra [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Add more friends to your messenger and enjoy! Go to http://messenger.yahoo.com/invite/ [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Goto loop
megh wrote: Is there anything like goto loop, which exists in most computer programs? e.g. I am looking for this kind of stuff : if(i 6) goto step-02 Any idea? Regards, It doesn't exist, but it can always be replaced by if() {} else {} constructs. (You don't usually see goto in the class of functional programming languages to which R belongs. See also http://en.wikipedia.org/wiki/Edsger_W._Dijkstra .) R does have break, next, and return() which cover some common cases where goto might be used with some reason. -- O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Row and Column positions
Hi R, m=data.frame(a=c(1,NA,5,5),b=c(4,5,6,7),c=c(NA,NA,NA,5)) I want to know the methods of getting row and column positions of NA in the above dataframe. How do I do this? Thanks, Shubha This e-mail may contain confidential and/or privileged i...{{dropped:13}} __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Row and Column positions
Am Freitag 31 Oktober 2008 12:17:30 schrieb Shubha Vishwanath Karanth: m=data.frame(a=c(1,NA,5,5),b=c(4,5,6,7),c=c(NA,NA,NA,5)) ? which HTH Claudia -- Claudia Beleites Dipartimento dei Materiali e delle Risorse Naturali Università degli Studi di Trieste Via Alfonso Valerio 6/a I-34127 Trieste phone: +39 (0 40) 5 58-34 47 email: [EMAIL PROTECTED] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] *** buffer overflow detected ***: /usr/lib64/R/bin/exec/R terminated on R 2.6.2 to 2.8.0: logging a bug?
Dear friends, there is a serious problem with RODBC on R from 2.6.2 to 2.8.0 on a postgresql 8.3 database. Let's say we define an empty dsn called test, with user postgres and password none. We test the connection with another application, and it works properly. When you call the database using RODBC, you get the following: odbcConnect(test,postgres,none) *** buffer overflow detected ***: /usr/lib64/R/bin/exec/R terminated === Backtrace: = /lib64/libc.so.6(__fortify_fail+0x37)[0x2ad38a3f4157] /lib64/libc.so.6[0x2ad38a3f27e0] /lib64/libc.so.6[0x2ad38a3f1db9] /lib64/libc.so.6(_IO_default_xsputn+0x8e)[0x2ad38a379b5e] /lib64/libc.so.6(_IO_vfprintf+0x3c22)[0x2ad38a351fa2] /lib64/libc.so.6(__vsprintf_chk+0x9d)[0x2ad38a3f1e5d] /lib64/libc.so.6(__sprintf_chk+0x80)[0x2ad38a3f1da0] /usr/lib64/libodbcpsql.so.2(SQLGetInfo+0x973)[0x2ad38de11013] /usr/lib64/libodbc.so.1(SQLGetInfo+0x624)[0x2ad38c5e91a4] /usr/lib64/R/library/RODBC/libs/RODBC.so(RODBCGetInfo+0xbb)[0x2ad38c393a9b] /usr/lib64/R/lib/libR.so[0x2ad389c2562d] /usr/lib64/R/lib/libR.so(Rf_eval+0x6a1)[0x2ad389c4ef31] /usr/lib64/R/lib/libR.so[0x2ad389c51dc3] /usr/lib64/R/lib/libR.so(Rf_eval+0x474)[0x2ad389c4ed04] /usr/lib64/R/lib/libR.so[0x2ad389c5014c] /usr/lib64/R/lib/libR.so(Rf_eval+0x474)[0x2ad389c4ed04] /usr/lib64/R/lib/libR.so(Rf_applyClosure+0x2aa)[0x2ad389c5268a] /usr/lib64/R/lib/libR.so(Rf_eval+0x38c)[0x2ad389c4ec1c] /usr/lib64/R/lib/libR.so[0x2ad389c5014c] /usr/lib64/R/lib/libR.so(Rf_eval+0x474)[0x2ad389c4ed04] /usr/lib64/R/lib/libR.so(Rf_applyClosure+0x2aa)[0x2ad389c5268a] /usr/lib64/R/lib/libR.so(Rf_eval+0x38c)[0x2ad389c4ec1c] /usr/lib64/R/lib/libR.so(Rf_ReplIteration+0x1c3)[0x2ad389c733c3] /usr/lib64/R/lib/libR.so(run_Rmainloop+0xd1)[0x2ad389c736d1] /usr/lib64/R/bin/exec/R(main+0x2b)[0x4008bb] /lib64/libc.so.6(__libc_start_main+0xf4)[0x2ad38a32a074] /usr/lib64/R/bin/exec/R[0x400769] === Memory map: 0040-00401000 r-xp 08:02 815595 /usr/lib64/R/bin/exec/R 0060-00601000 rw-p 08:02 815595 /usr/lib64/R/bin/exec/R 00601000-01bf3000 rw-p 00601000 00:00 0 [heap] 2ad38995c000-2ad389976000 r-xp 08:02 2044002/lib64/ld-2.7.so 2ad389976000-2ad389978000 rw-p 2ad389976000 00:00 0 2ad389b75000-2ad389b76000 r--p 00019000 08:02 2044002/lib64/ld-2.7.so 2ad389b76000-2ad389b77000 rw-p 0001a000 08:02 2044002/lib64/ld-2.7.so 2ad389b77000-2ad389e03000 r-xp 08:02 878425 /usr/lib64/R/lib/libR.so 2ad389e03000-2ad38a002000 ---p 0028c000 08:02 878425 /usr/lib64/R/lib/libR.so 2ad38a002000-2ad38a016000 rw-p 0028b000 08:02 878425 /usr/lib64/R/lib/libR.so 2ad38a016000-2ad38a0af000 rw-p 2ad38a016000 00:00 0 2ad38a0af000-2ad38a0db000 r-xp 08:02 878426 /usr/lib64/R/lib/libRblas.so 2ad38a0db000-2ad38a2da000 ---p 0002c000 08:02 878426 /usr/lib64/R/lib/libRblas.so 2ad38a2da000-2ad38a2db000 rw-p 0002b000 08:02 878426 /usr/lib64/R/lib/libRblas.so 2ad38a2db000-2ad38a2dc000 r--p 08:02 892499 /usr/share/locale/en_GB.UTF-8/LC_MESSAGES/SYS_LC_MESSAGES 2ad38a2dc000-2ad38a2e3000 r--s 08:02 654700 /usr/lib64/gconv/gconv-modules.cache 2ad38a30b000-2ad38a30c000 rw-p 2ad38a30b000 00:00 0 2ad38a30c000-2ad38a457000 r-xp 08:02 2044009/lib64/libc-2.7.so 2ad38a457000-2ad38a656000 ---p 0014b000 08:02 2044009/lib64/libc-2.7.so 2ad38a656000-2ad38a65a000 r--p 0014a000 08:02 2044009/lib64/libc-2.7.so 2ad38a65a000-2ad38a65b000 rw-p 0014e000 08:02 2044009/lib64/libc-2.7.so 2ad38a65b000-2ad38a66 rw-p 2ad38a65b000 00:00 0 2ad38a66-2ad38a718000 r-xp 08:02 2396466/usr/lib64/libgfortran.so.2.0.0 2ad38a718000-2ad38a917000 ---p 000b8000 08:02 2396466/usr/lib64/libgfortran.so.2.0.0 2ad38a917000-2ad38a919000 rw-p 000b7000 08:02 2396466/usr/lib64/libgfortran.so.2.0.0 2ad38a919000-2ad38a998000 r-xp 08:02 2044017/lib64/libm-2.7.so 2ad38a998000-2ad38ab98000 ---p 0007f000 08:02 2044017/lib64/libm-2.7.so 2ad38ab98000-2ad38ab99000 r--p 0007f000 08:02 2044017/lib64/libm-2.7.so 2ad38ab99000-2ad38ab9a000 rw-p 0008 08:02 2044017/lib64/libm-2.7.so 2ad38ab9a000-2ad38ab9b000 rw-p 2ad38ab9a000 00:00 0 2ad38ab9b000-2ad38abd1000 r-xp 08:02 2044184/lib64/libreadline.so.5.2 2ad38abd1000-2ad38add ---p 00036000 08:02 2044184/lib64/libreadline.so.5.2 2ad38add-2ad38add8000 rw-p 00035000 08:02 2044184/lib64/libreadline.so.5.2 2ad38add8000-2ad38add9000 rw-p 2ad38add8000 00:00 0
[R] Goto loop
Is there anything like goto loop, which exists in most computer programs? e.g. I am looking for this kind of stuff : if(i 6) goto step-02 Any idea? Regards, -- View this message in context: http://www.nabble.com/%22Goto%22-loop-tp20263445p20263445.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Goto loop
2008/10/31 megh [EMAIL PROTECTED]: Is there anything like goto loop, which exists in most computer programs? Really? Not since 1968, I think: http://www.cs.utexas.edu/users/EWD/ewd02xx/EWD215.PDF e.g. I am looking for this kind of stuff : if(i 6) goto step-02 Any idea? Thinking you need a 'goto' is a massive red flag in your program. This red flag is waving to warn you of danger ahead. As well as Dijkstra's somewhat technical reasons for abhoring such a language construct, it will also make your program very difficult to understand. R has for-loops, repeat-loops, while loops, and conditional (if-then-else) structures. You can break out of loops with 'break', and jump to the next iteration with 'next'. There is no goto: help.search(goto) No help files found with alias or concept or title matching 'goto' using regular expression matching. If you have been given a bolognese full of spaghetti code (which is what we call anything with a generous sprinkling of goto statements) and you want to rewrite it in R, you'll have some thinking to do. Or just write it in Fortran which, since it dates from well before Dijkstra's famous note, has GOTO statements you can use. Barry __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] how to retrieve some records from a dataframe
See head: head(Df, n = 30) On Fri, Oct 31, 2008 at 6:44 AM, Kurapati, Ravichandra (Ravichandra) [EMAIL PROTECTED] wrote: Hi , Dataframe Df contains more than 30 no of records.but I want only first 30 rows of data. How can I retrieve it. Thanks K.Ravichandra [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.htmlhttp://www.r-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Henrique Dallazuanna Curitiba-Paraná-Brasil 25° 25' 40 S 49° 16' 22 O [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Row and Column positions
Try: which(is.na(m), arr = T) On Fri, Oct 31, 2008 at 9:17 AM, Shubha Vishwanath Karanth [EMAIL PROTECTED] wrote: Hi R, m=data.frame(a=c(1,NA,5,5),b=c(4,5,6,7),c=c(NA,NA,NA,5)) I want to know the methods of getting row and column positions of NA in the above dataframe. How do I do this? Thanks, Shubha This e-mail may contain confidential and/or privileged i...{{dropped:13}} __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.htmlhttp://www.r-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Henrique Dallazuanna Curitiba-Paraná-Brasil 25° 25' 40 S 49° 16' 22 O [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] rimage don't install on mac OS X 10.5.5
Thanks for your rapid response. I tried your sugerences and set C_INCLUDE_PATH=/usr/local/include. I also look in /usr/local/include and here it appears fftw.h and also fftw3.f, fftw3.h, fftw_threads.h and rfftw.h rfftw_threads.h but it don't work also. I will try to update to R 2.8.0 Thanks for your sugestions El 31/10/2008, a las 10:10, Prof Brian Ripley escribió: On Fri, 31 Oct 2008, Florencio Gonzalez Blazquez wrote: Hello, i'm trying to install de rimage package form source. I have installed fftw2.1.5 and libjpeg, but during compilation appear this message: Where did you install the fftw headers? If in /usr/local/include, this is apparently not in the C_INCLUDE_PATH for your system. (The reason that configure finds ffw.h but the compile does not is most likely that -isysroot has been set in the compile: the author of rimage has not been careful to use the same compiler flags as R uses. Cc:ing the author here -- please study 'Writing R Extensions' about how to use a configure script in a package.) I suspect the workaround is to set C_INCLUDE_PATH=/usr/local/include (or as appropriate) in the environment used for the compile, or to install ffw with --prefix=/usr. There is a list, R-sig-mac, for MacOS-specific questions and you will get the ears of the MacOS experts there. Please note the advice of the posting guide to update your R *before posting*: R 2.7.0 is rather old and many bugs have been fixed since. * Installing *source* package 'rimage' ... checking for g++... g++ checking for C++ compiler default output... a.out checking whether the C++ compiler works... yes checking whether we are cross compiling... no checking for suffix of executables... checking for suffix of object files... o checking whether we are using the GNU C++ compiler... yes checking whether g++ accepts -g... yes checking for gcc... gcc checking whether we are using the GNU C compiler... yes checking whether gcc accepts -g... yes checking for gcc option to accept ANSI C... none needed checking how to run the C preprocessor... gcc -E checking for egrep... grep -E checking for ANSI C header files... yes checking for sys/types.h... yes checking for sys/stat.h... yes checking for stdlib.h... yes checking for string.h... yes checking for memory.h... yes checking for strings.h... yes checking for inttypes.h... yes checking for stdint.h... yes checking for unistd.h... yes checking fftw.h usability... yes checking fftw.h presence... yes checking for fftw.h... yes checking jpeglib.h usability... yes checking jpeglib.h presence... yes checking for jpeglib.h... yes checking for inline... inline checking for stdlib.h... (cached) yes checking for GNU libc compatible malloc... yes configure: creating ./config.status config.status: creating src/Makevars ** libs ** arch - i386 g++ -arch i386 -isysroot /Developer/SDKs/MacOSX10.4u.sdk -mmacosx- version-min=10.4 -I/Library/Frameworks/R.framework/Resources/ include -I/Library/Frameworks/R.framework/Resources/include/i386 -g -O2 -msse3 -fPIC -g -O2 -march=nocona -c equalize.cpp -o equalize.o gcc -arch i386 -isysroot /Developer/SDKs/MacOSX10.4u.sdk -mmacosx- version-min=10.4 -std=gnu99 -I/Library/Frameworks/R.framework/ Resources/include -I/Library/Frameworks/R.framework/Resources/ include/i386 -g -O2 -msse3 -fPIC -g -O2 -march=nocona -c fftw_access_func.c -o fftw_access_func.o fftw_access_func.c:10:18: error: fftw.h: No such file or directory fftw_access_func.c: In function ‘fftw_access_func’: fftw_access_func.c:20: error: ‘fftwnd_plan’ undeclared (first use in this function) fftw_access_func.c:20: error: (Each undeclared identifier is reported only once fftw_access_func.c:20: error: for each function it appears in.) fftw_access_func.c:20: error: syntax error before ‘plan’ fftw_access_func.c:21: error: ‘fftw_complex’ undeclared (first use in this function) fftw_access_func.c:21: error: ‘i’ undeclared (first use in this function) fftw_access_func.c:21: error: ‘s’ undeclared (first use in this function) fftw_access_func.c:32: error: syntax error before ‘)’ token fftw_access_func.c:33: error: syntax error before ‘)’ token fftw_access_func.c:39: error: ‘fftw_real’ undeclared (first use in this function) fftw_access_func.c:39: error: syntax error before ‘img’ fftw_access_func.c:40: error: syntax error before ‘img’ fftw_access_func.c:43: error: ‘plan’ undeclared (first use in this function) fftw_access_func.c:43: warning: implicit declaration of function ‘fftw2d_create_plan’ fftw_access_func.c:43: error: ‘FFTW_ESTIMATE’ undeclared (first use in this function) fftw_access_func.c:45: warning: implicit declaration of function ‘fftwnd_one’ fftw_access_func.c:46: warning: implicit declaration of function ‘fftwnd_destroy_plan’ make: *** [fftw_access_func.o] Error 1 chmod: /Library/Frameworks/R.framework/Resources/library/rimage/ libs/i386/*: No such file or directory ** Removing
[R] dotplot with a log scale
Dear all, I would like to create a dotplot (lattice) with a log scale as the function http://www.nabble.com/dotchart-with-log-scale--td5411625.html#a5411625 mydotchart of Johannes Huesing. Is it possible? Thanx Gianandrea -- View this message in context: http://www.nabble.com/dotplot-with-a-log-scale-tp20265718p20265718.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Fitting weibull and exponential distributions to left censoring data
Use the survreg function. There are many different ways to parameterize a Weibull. The survreg function imbeds it a general location-scale familiy, which is a different parameterization than the rweibull function. y - rweibull(1000, shape=2, scale=5) survreg(Surv(y)~1, dist=weibull) Coefficients: (Intercept) 1.592543 Scale= 0.5096278 Loglik(model)= -2201.9 Loglik(intercept only)= -2201.9 survreg's scale =1/(rweibull shape) survreg's intercept = log(rweibull scale) For the log-likelihood all parameterizations lead to the same value. There is not right or wrong parameterization for a Weibull (IMHO), but there certainly is a lot of room for confusion. This comes up enough that I have just added it as an example in the survreg help page, which will migrate to the general R distribution in due course. Terry Therneau __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] locfit and latest R version
I have been using some old R scripts that were prepared for me using R Version 1.9.0. In these, there is a call to library(locfit) and with this version of R, I have no problem. But needing to upgrade to the latest version of R, I find these scripts no longer work, and with help(locfit) I get Error in help(locfit): Object locfit not found So my question is - what is available that has replaced locfit, and that will do the same thing as locfit, but with the latest R version? You will realise from this that I have no expertise in using R or in writing R scripts, and will greatly appreciate any advice. Many thanks Stuart Barker J.S.F. (Stuart) Barker HonDSc FTSE Emeritus Professor School of Environmental and Rural Science University of New England Armidale, NSW 2351 Australia Honorary Professor School of Integrative Biology University of Queensland Brisbane HOME: 114 Cooke Road Witta, Maleny Qld 4552 Ph. HOME ++ 61 7 5435 8365 email - [EMAIL PROTECTED] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] PAM: how to get the best number of clusters
My problem is that I have already a distance (similarity) matrix generated outside R through a C++ code because the criteria to calculate the distance between pairs of objects are none of the standard criteria implemented in R. If I got it right, but I might be mistaken, stepFlexclust() performs the clusters layout optimization by calling either one of kcca or cclust which calculate their own similarity matrix. I just need a function or method to optimize the clusters number no matter how the similarity matrix has been generated and no matter which clustering function I use (PAM). Is this at all possible ? Thank you very much, Maura On Fri, Oct 31, 2008 at 12:18 AM, Dylan Beaudette [EMAIL PROTECTED] wrote: On Thursday 30 October 2008, Maura E Monville wrote: I have the book you mentioned. It basically describes the silhouette method. I do not have it handy as I moved so it is still in some box. However I cannot remember that book providing any other criterion to find the best clusters number. On the other hand I have the same problem with hierarchical clustering techniques. I use clusters as exploratory analysis because I do not have any a-priori knowledge that helps me make a choice. How can multivariate analysis help? I launched a loop where the silhouette test follows PAM which is passed a clusters number increased by 1 at each iteration. Since I am observing that the silhouette value is now oscillating among negative numbers, I wonder whether I can assume that it can only grow worse once it has turned negative the first time so leave the loop after the first negative number and choose the clusters number associated with the biggest positive silhouette value. This procedure would spare a lot of CPU time. Another approach might involve the stepFlexclust() from the flexclust package. See the manual page for this function for examples. Dylan Thank you very much, Maura On Thu, Oct 30, 2008 at 7:25 PM, Dylan Beaudette [EMAIL PROTECTED]wrote: On Thursday 30 October 2008, Maura E Monville wrote: I have a pretty big similarity matrix (2870x2870). I will produce even bigger ones soon. I am using PAM to generate clusters. The desired number of output clusters is a PAM input parameter. I do not know a-priopri what is the best clusters layout . I resorted to the silhouette test. It takes forever as I have to run PAM with all possible numbers of clusters. I wonder whether there is some faster method, either a s/w code or some theoretical guidelines, to get the optimum clusters number. Thank you very much, This is a very general topic in the field of multivariate analysis. There really isn't any way to know the 'correct' number of clusters, however there are several metrics that can give you an indication of how messy your data are. For information on the methods in the cluster package, see this book: Kaufman, L. Rousseeuw, P. J. Finding Groups in Data An Introduction to Cluster Analysis Wiley-Interscience, 2005 Otherwise, consider a book on multivariate analysis. Alternatively, try a hierarchical clustering approach, and look for meaningful groupings. Some thing like this: d - diana(daisy(your_data_matrix)) d.hc - as.hclust(d) d.hc$labels - your_data_matrix$id plot(d.hc) Cheers, Dylan -- Dylan Beaudette Soil Resource Laboratory http://casoilresource.lawr.ucdavis.edu/ University of California at Davis 530.754.7341 -- Dylan Beaudette Soil Resource Laboratory http://casoilresource.lawr.ucdavis.edu/ University of California at Davis 530.754.7341 -- Maura E.M [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] chi-square area analysis
Hi I was wondering if anyone have done an area analysis using chi-aquare? I have three areas (area1, area2 and area3) I need to compare them to each other by looking at a binomial distribution factor (0=negative, 1=positive). I have already done a logistiv regression analysis comparing these areas, but I have been told to do a chi-square, but I cannot find any commands for area comparing. Thanks for your help! - Arctica -- View this message in context: http://www.nabble.com/chi-square-area-analysis-tp20266619p20266619.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] how to compute a roc curve
Hi, I'm trying to set up a prediction software, now i testing the performance of my method, so i need to calculate a ROC curve, specially auc, cut-off, sens and spec, i just looking at ROCH package, but it's a mass for me, i'm not a math guy and I'm getting lost Could any of you recommend me an easy-to-use package to do this task? i just have a list of positive/negative samples and his score on my program. can I compute a roc curve with this? thanks pau -- Pau Marc Muñoz Torres Laboratori de Biologia Computacional Institut de Biotecnologia i Biomedicina Vicent Villar Universitat Autonoma de Barcelona E-08193 Bellaterra (Barcelona) telèfon: 93 5812807 Email : [EMAIL PROTECTED] [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] stratified kappa (measure agreement or interrater reliability)?
On Oct 30, 2008, at 11:38 PM, FD wrote: Hi All: Could anyone point me to a package that can calculate stratified kappa? My design is like this, 4 raters, 30 types of diagnosis scores, 20 patients. Each rater will rate each patient for each type of diagnosis score. The rater's value is nominal. I know I can measure the agreement between raters for each type of diagnosis score, e.g., calculate out 30 kappa values. My problem is I want to have an overall agreement measure (a single value and its significance over chance). Could anyone help me with this? I am not a statistician or a psychometrician, so have no experience with any of these packages. A google search on produced this link: http://www.mail-archive.com/[EMAIL PROTECTED]/ msg89858.html ... and looking in package psy in CRAN, I see lkappa(), Light’s kappa for n raters, which seems to meet your specifications. The concord package may have the facilities but I am not able to tell from the documentation. Perhaps Jim Lemon can be queried. -- David Winsemius __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] locfit and latest R version
My guess is that you have not configured you new installation to see (and update) your collection of packages. As a quick fix, what happens when you go to your favorite CRAN mirror and install a new copy of locfit? My version is 1.5-4 and runs with recent versions of R. -- David Winsemius, MD Heritage Labs On Oct 31, 2008, at 7:21 AM, Stuart Barker (Prof) wrote: I have been using some old R scripts that were prepared for me using R Version 1.9.0. In these, there is a call to library(locfit) and with this version of R, I have no problem. But needing to upgrade to the latest version of R, I find these scripts no longer work, and with help(locfit) I get Error in help(locfit): Object locfit not found So my question is - what is available that has replaced locfit, and that will do the same thing as locfit, but with the latest R version? You will realise from this that I have no expertise in using R or in writing R scripts, and will greatly appreciate any advice. Many thanks Stuart Barker J.S.F. (Stuart) Barker HonDSc FTSE Emeritus Professor School of Environmental and Rural Science University of New England Armidale, NSW 2351 Australia Honorary Professor School of Integrative Biology University of Queensland Brisbane HOME: 114 Cooke Road Witta, Maleny Qld 4552 Ph. HOME ++ 61 7 5435 8365 email - [EMAIL PROTECTED] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] subset with variable number of subset arguments
I think subset(set, InfCode %in% x) does it, where 'x' is your vector of ids (e.g., c(1,3,9) in your example at the bottom) Brar Piening wrote: I've got a big data.frame from which I need to extract data based on a variable number of Id's (InfCode). Until now I've been using the following dull solution as I never needed to search for more than 5 codes. Now my needs have increased faster than my R skills did and I need to call my function with about 25 values for x. There has to be a *apply or even simpler solution which (after RTM) I'm not able to figure out. Can anybody direct me to a possible solution? Regards, Brar GetInfCodeSubset - function(set, x){ if(1 == length(x)) return(subset(set, (set$InfCode == x[1]))) else if(2 == length(x)) return(subset(set, (set$InfCode == x[1]) | (set$InfCode == x[2]))) else if(3 == length(x)) return(subset(set, (set$InfCode == x[1]) | (set$InfCode == x[2]) | (set$InfCode == x[3]))) else if(4 == length(x)) return(subset(set, (set$InfCode == x[1]) | (set$InfCode == x[2]) | (set$InfCode == x[3]) | (set$InfCode == x[4]))) else if(5 == length(x)) return(subset(set, (set$InfCode == x[1]) | (set$InfCode == x[2]) | (set$InfCode == x[3]) | (set$InfCode == x[4]) | (set$InfCode == x[5]))) else if(6 == length(x)) return(subset(set, (set$InfCode == x[1]) | (set$InfCode == x[2]) | (set$InfCode == x[3]) | (set$InfCode == x[4]) | (set$InfCode == x[5]) | (set$InfCode == x[6]))) else stop(Too many elements in x) } set - data.frame(PatId = c(1:100), InfCode = rep(c(1:20), 5)) GetInfCodeSubset(set, c(1,3,9)) -- __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] getting the p-value from lm as a list object
Hi, I'm trying to get the p-value from the 'lm' regression function as a list object. For example, I can get r^2 from the following code by entering summary(fm)$r.squared. Is there a way to get the p-value? If not, is there a function where I can enter the f-value and degrees of freedom to get the p-value? Thanks. x - c(1,2,3,4,5,6,7,8,9,10) y - c(1,2,3,4,4,5,6,8,1,9) fm - lm(y ~ x) str(summary(fm)) [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] how to compute a roc curve
Try library(ROCR) Pau Marc Munoz Torres wrote: Hi, I'm trying to set up a prediction software, now i testing the performance of my method, so i need to calculate a ROC curve, specially auc, cut-off, sens and spec, i just looking at ROCH package, but it's a mass for me, i'm not a math guy and I'm getting lost Could any of you recommend me an easy-to-use package to do this task? i just have a list of positive/negative samples and his score on my program. can I compute a roc curve with this? thanks pau __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] getting the p-value from lm as a list object
eric lee wrote: Hi, I'm trying to get the p-value from the 'lm' regression function as a list object. For example, I can get r^2 from the following code by entering summary(fm)$r.squared. Is there a way to get the p-value? If not, is there a function where I can enter the f-value and degrees of freedom to get the p-value? Thanks. x - c(1,2,3,4,5,6,7,8,9,10) y - c(1,2,3,4,4,5,6,8,1,9) fm - lm(y ~ x) str(summary(fm)) What about the following (taken from stats:::print.summary.lm): x - c(1,2,3,4,5,6,7,8,9,10) y - c(1,2,3,4,4,5,6,8,1,9) fm - lm(y ~ x) summary(fm) # for comparison only sfm - summary(fm) pf(sfm$fstatistic[1], sfm$fstatistic[2], sfm$fstatistic[3], lower.tail = FALSE) Thomas P. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] getting the p-value from lm as a list object
on 10/31/2008 10:59 AM eric lee wrote: Hi, I'm trying to get the p-value from the 'lm' regression function as a list object. For example, I can get r^2 from the following code by entering summary(fm)$r.squared. Is there a way to get the p-value? If not, is there a function where I can enter the f-value and degrees of freedom to get the p-value? Thanks. x - c(1,2,3,4,5,6,7,8,9,10) y - c(1,2,3,4,4,5,6,8,1,9) fm - lm(y ~ x) str(summary(fm)) The default output is created in stats:::print.summary.lm(). The basic incantation is: pf(x$fstatistic[1], x$fstatistic[2], x$fstatistic[3], lower.tail = FALSE) where: pf() is the F distribution function x = the summary.lm model object x$fstatistic[1] = model F statistic x$fstatistic[2] = model numerator DF x$fstatistic[3] = model denominator DF Thus, using lm.D9 from example(lm): summary(lm.D9) Call: lm(formula = weight ~ group) Residuals: Min 1Q Median 3Q Max -1.0710 -0.4938 0.0685 0.2462 1.3690 Coefficients: Estimate Std. Error t value Pr(|t|) (Intercept) 5.0320 0.2202 22.850 9.55e-15 *** groupTrt -0.3710 0.3114 -1.1910.249 --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 0.6964 on 18 degrees of freedom Multiple R-squared: 0.07308,Adjusted R-squared: 0.02158 F-statistic: 1.419 on 1 and 18 DF, p-value: 0.249 summary(lm.D9)$fstatistic value numdf dendf 1.419101 1.00 18.00 pf(1.419, 1, 18, lower = FALSE) [1] 0.2490394 See ?pf HTH, Marc Schwartz __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Help needed with Waterfall plot
Hi friends, I need suggestions/directions on how to producing a waterfall plot for present extend of change in tumour size for a set of respondents in a study. Example of use of waterfall plot is in the following slides presented at ASCO 2007 by Axel Grothey. Link is http://media.asco.org/player/default.aspx?LectureID=AG265conferenceFolder=GI2007SessionFolder=Posterslideonly=yesTrackID=N929LectureTitle=Waterfall%20plots%20provide%20detailed%20information%20on%20magnitude%20of%20response%20to%20conventional%20chemotherapy%20in%20colorectal%20cancer%3a%20Lessons%20learned%20from%20N9741.Key=vm_45_3_26_265SpeakerName=%3b%20Presenter%3a%20Axel%20Grothey%2c%20MDmediaURL=%2fmediaServerName=media.asco.orgmax=12ext=jpguseASX=falseplaytype=playtype=playtype=, The link is pretty long but it takes you right to the presentation. A Smile costs Nothing But Rewards Everything Happiness is not perfected until it is shared -Jane Porter [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] set the language
On Fri, 31 Oct 2008, Gustave Lefou wrote: Hello, I would like to change my R language from German to French or English. On Windows, we presume. I have read messages in the archives and also R Installation and Administration paragraph 7.2 but I did not get it all. I have to set LANGUAGE=en (what would it be for French, fr ?), but I don't know where. Well, the rw-FAQ does tell you where: 1. On the command line as name=value pairs. For example in the shortcut to `Rgui' you could have path_to_R\bin\Rgui.exe HOME=p:/ R_LIBS=p:/myRlib so in your shortcut use LANGUAGE=fr . (That's the ISO 639 abbreviation for French, and the manual does tell you that is what to use! Google very rapidly produces http://en.wikipedia.org/wiki/ISO_639-1 as a list.) That's the best place as it sets the language only for R. I am under Windows XP. FAQ 2.15 tells me : under Windows 2000/XP/2003 you can use `System' in the control panel or the properties of `My Computer' (under the `Advanced' tab). Is that right ? I have problem finding all this, partly because my Windows is in German. I give you the German names I have found : Systemsteuerung -- System -- Umgebungsvariablen Should I click on Neu (Neu Benutzervariable) and then enter LANGUAGE as Name and en as Wert ? Maybe I am completely wrong. Thank you very much [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Help needed with Waterfall plot
Philip Twumasi-Ankrah schrieb: Hi friends, I need suggestions/directions on how to producing a waterfall plot for present extend of change in tumour size for a set of respondents in a study. Example of use of waterfall plot is in the following slides presented at ASCO 2007 by Axel Grothey. Link is http://media.asco.org/player/default.aspx?LectureID=AG265conferenceFolder=GI2007SessionFolder=Posterslideonly=yesTrackID=N929LectureTitle=Waterfall%20plots%20provide%20detailed%20information%20on%20magnitude%20of%20response%20to%20conventional%20chemotherapy%20in%20colorectal%20cancer%3a%20Lessons%20learned%20from%20N9741.Key=vm_45_3_26_265SpeakerName=%3b%20Presenter%3a%20Axel%20Grothey%2c%20MDmediaURL=%2fmediaServerName=media.asco.orgmax=12ext=jpguseASX=falseplaytype=playtype=playtype=, The link is pretty long but it takes you right to the presentation. Hi Phillip, is this a waterfall plot: ## a few data x - 0:99 y - sort(rnorm(100), decreasing=TRUE) # the plot plot(y, type=n) polygon(c(min(x), x, max(x), 0), c(0, y, 0, 0), col=green) Thomas P. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Question about betareg using R
Good morning, I am using R to try to model the proportion of burned area in Portugal. The dependent variable is the proportion (P). The family used is binomial and the epsilon would be binary. They have recommended me to use the Betareg package as it allows to work with proportions. When I use the Betareg function for my data (2800 lines) with 7 variavels. I created a model using the sum of all seven variables and using p (proportion of area burnt as the independent variable). I obtained these results, can you tell me if this is normal? or what do the errors mean? Thank you in advance. Is there any limit in the number of lines and variables(columns) to use? This was the functions used and the results obtained: attach(dados) mod9 - betareg(P ~ Fuel+Temperature+Slope+Altitude+Precipitation+Roadsproximity+Population, data=dados) Warning messages: 1: NaNs produced in: sqrt(W) 2: NaNs produced in: sqrt(W) 3: NaNs produced in: sqrt(1 + phihat) summary(mod9) Call: betareg(formula = P ~ Fuel + Temperature + Slope + Altitude + Precipitation + Roadsproximity + Population, data = dados) DevianceResiduals: Min.1st Qu. Median 3rd 0 0 0 0 0 Coefficients: EstimateStd.Error z-value Pr(|z|) (Intercept) -83062.59.34E-02-8.89E+05 0 FuelAnnualCrop -2256.8 2.31E-01-9.78E+03 0 FuelHard -8179.8 7.95E-02-1.03E+05 0 FuelHardSoftEuc -4449.7 7.02E-02-6.34E+04 0 FuelNoFuel -9552.1 9.02E-02-1.06E+05 0 FuelPermCrop-5112.5 8.39E-02-6.10E+04 0 FuelShrubs -8246 7.32E-02-1.13E+05 0 FuelSoft -11795.17.51E-02-1.57E+05 0 FuelSoftEuc -1561.8 7.67E-02-2.04E+04 0 Temperature -1.3E+077.14E-04-1.80E+10 0 Slope 78122.8 3.78E-032.07E+070 Altitude 1143023 7.19E-051.59E+100 Precipitation -14822882.11E-03-7.03E+08 0 RoadsproximityDistRd1Km-45315.13.82E-02-1.19E+06 0 Population 4711684 2.81E-041.68E+100 Estimated precision parameter (phi): -560353.4 with s.e. 315.9113 Null Deviance: 0 on 2818 degrees of freedom Residual Deviance: 0 on 2804 degrees of freedom Log-Likelihood Ratio Statistic: 3291846935 on 14 degrees of freedom Pseudo R^2: 0.07753642 AIC: -3291849573 Warning messages: 1: NaNs produced in: sqrt(muhat * (1 - muhat)/(1 + phihat)) 2: NaNs produced in: sqrt(muhat * (1 - muhat)/(1 + phihat)) Thank you very much in advance, Jordi -- __ Jordi Garcia-Gonzalo FORCHANGE - Forest Ecosystem Management Under Global Change Departamento de Engenharia Florestal - Instituto Superior de Agronomia Universidade Técnica de Lisboa Tapada da Ajuda, 1349-017 Lisboa Portugal tel: +351 21 365 33 66 fax: +351 21 364 50 00 email:[EMAIL PROTECTED] [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Quantile Regression for Longitudinal Data:error message
Quantile Regression for Longitudinal Data. Hi, I am trying to estimate a quantile regression using panel data. I am trying to use the model that is described in Dr. Koenker's article. So I use the code the that is posted in the following link: http://www.econ.uiuc.edu/~roger/research/panel/rq.fit.panel.R I am trying to change the number quantiles being estimated. I change the codes about w and taus ,as following w=c(.01,.02,.03,.04,.05,.06,.07,.08,.09,.1,.11,.12,.13,.14,.15,.16,.17 ,.18,.19,.2,.21,.22,.23,.24,.25,.26,.27,.28,.29,.3,.31,.32,.33,.34,.35 ,.36,.37,.38,.39,.4,.41,.42,.43,.44,.45,.46,.47,.48,.49,.5,.49,.48,.47 ,.46,.45,.44,.43,.42,.41,.4,.39,.38,.37,.36,.35,.34,.33,.32,.31,.3,.29 ,.28,.27,.26,.25,.24,.23,.22,.21,.2,.19,.18,.17,.16,.15,.14,.13,.12,.11 ,.1,.09,.08,.07,.06,.05,.04,.03,.02,.01) ,taus=(1:99)/100,lambda = 1 But I get error message: .local(x, pivot, ...) : Increase tmpmax So I am wondering if I am doing something wrong or I mistake w's meaning. Thanks I really would appreciate some suggestions. Best Helen Chen -- View this message in context: http://www.nabble.com/Quantile-Regression-for-Longitudinal-Data%3Aerror-message-tp20269386p20269386.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Kalman Filter
Hi, I am studying Kalman Filter and it seems to be difficult for me to apply the filter on a simple ARMA. It is easy to construct the state-space model, for instance: dlmModARMA(ar=c(0.4,-0.2),ma=c(0.2,-0.1, sigma2=1) but applying the dlmFilter on it, it doesn't work... I don't know if my problem is clear but if anyone has already worked on Kalman filter, it could be great to advise me! Thank you in advance! Sandrine __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Goto loop
2008/10/31 Peter Dalgaard [EMAIL PROTECTED]: It doesn't exist, but it can always be replaced by if() {} else {} constructs. (You don't usually see goto in the class of functional programming languages to which R belongs. See also http://en.wikipedia.org/wiki/Edsger_W._Dijkstra .) Also see also the always wonderful xkcd comic on the subject: http://xkcd.com/292/ Barry __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] [ifelse] how to maintain a value from original matrix without probs?
A - matrix(-4:4, ncol=3) A [,1] [,2] [,3] [1,] -4 -12 [2,] -303 [3,] -214 Apos - A; Apos[A = 0] - NA; Apos [,1] [,2] [,3] [1,] NA NA2 [2,] NA NA3 [3,] NA14 Aneg - A; Aneg[A = 0] - NA Aneg [,1] [,2] [,3] [1,] -4 -1 NA [2,] -3 NA NA [3,] -2 NA NA /Henrik On Fri, Oct 31, 2008 at 7:59 AM, Diogo André Alagador [EMAIL PROTECTED] wrote: Dear all, I have a matrix with positive and negative values. From this I would like to produce 2 matrices: 1st - retaining positives and putting NA in other positions 2nd - retaining negatives and putting NA in other positions and then apply rowMeans for both. I am trying to use the function ifelse in the exemplified form: ifelse(A0,A,NA) but by putting A as a 2nd parameter it changes dimensions of the original object. I wonder if I can do this, as it seems not to difficult. thanks in advance [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] [ifelse] how to maintain a value from original matrix without probs?
Hello - Diogo André Alagador wrote: Dear all, I have a matrix with positive and negative values. From this I would like to produce 2 matrices: 1st - retaining positives and putting NA in other positions 2nd - retaining negatives and putting NA in other positions and then apply rowMeans for both. I am trying to use the function ifelse in the exemplified form: ifelse(A0,A,NA) but by putting A as a 2nd parameter it changes dimensions of the original object I cannot reproduce this: a - matrix(sample(-10:10, 100, replace = TRUE), nrow = 10) ifelse(a 0 , a, NA) gives me a 10 x 10 matrix of positive values and NA values, with 10 x 10 being the original dimensions of a. rowMeans(ifelse(a 0 , a, NA), na.rm = TRUE) gives me what you'd like for positive values, I think? I wonder if I can do this, as it seems not to difficult. thanks in advance [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] [ifelse] how to maintain a value from original matrix without probs?
The ifelse function sometimes has unwanted side effects. Your problem can easily be done with simple subsetting. Try: pos - neg - A pos[ pos = 0 ] - NA neg[ neg = 0 ] - NA # remove the ='s if you don't want strict positive/negative Hope this helps, -- Gregory (Greg) L. Snow Ph.D. Statistical Data Center Intermountain Healthcare [EMAIL PROTECTED] 801.408.8111 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] project.org] On Behalf Of Diogo André Alagador Sent: Friday, October 31, 2008 8:59 AM To: r-help@r-project.org Subject: [R] [ifelse] how to maintain a value from original matrix without probs? Dear all, I have a matrix with positive and negative values. From this I would like to produce 2 matrices: 1st - retaining positives and putting NA in other positions 2nd - retaining negatives and putting NA in other positions and then apply rowMeans for both. I am trying to use the function ifelse in the exemplified form: ifelse(A0,A,NA) but by putting A as a 2nd parameter it changes dimensions of the original object. I wonder if I can do this, as it seems not to difficult. thanks in advance [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting- guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] R-WinEdt unexcpected Asian characters
Dear list! I have encountered an odd problem with R-WinEdt. The program is mostly running fine and I had no problems with the installation. But several times a day I'm typing in R-WinEdt there appears a menu with Chinese characters where the curser is and I'm not able to type normally. I dont know any other way to stop it except restarting R-WinEdt. After that it works normally for a while until it happens again. I was not able to figure out when it happens. Sometimes after error messages in R, but not always and I am never aware of having typed a certain key combination or anything like that. Is there anything I can change in the settings to prevent this from happening? I'm using R 2.8.0. and WinEdt 5.5. in Windows XP. My regional options in Windows are set to Sweden and Swedish. With earlier versions of R and WinEdt I did not have this problem. However, this is a different computer. Thank you for your help! Saskia __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] smooth function advice
Thanks to Mark Leeds for an implementation of exponentially weighted moving averages as follows which solves this problem. This also resmooths recursively. ewma-function(x,lambda = .5, init = (1-lambda)*.raw[good.ind][1],order=1) { .raw - unclass(coredata(x)) good.ind - !is.na(.raw) # determine good values if(order1) ewma(ewma(x,lambda,init,order=1),lambda,init,order=order-1) else { # work with 'non-zoo' data for speed and then recombine .raw[good.ind] - filter(lambda * .raw[good.ind], filter=(1-lambda), method='recursive',init=coredata(init)) zoo(.raw, index(x)) # create zoo object for return } } Tolga I Uzuner/JPMCHASE 30/10/2008 15:36 To r-help@r-project.org cc Subject smooth function advice Dear R Users, I am looking for a smoothing function with the following characteristics for a time series of data: - at each date, should only use data up to that date (so, right aligned and not centered) - should return a smoothed series of length equal to the original time series: - for a one-day time series, just returns that day - this means the front part of the series will not be as smooth... that's ok - if the original time series has length 10, the returned smoothed time series has length 10 - there should be some control over smoothness Any suggestions for a specific approach/package in R ? Thanks, Tolga Generally, this communication is for informational purposes only and it is not intended as an offer or solicitation for the purchase or sale of any financial instrument or as an official confirmation of any transaction. In the event you are receiving the offering materials attached below related to your interest in hedge funds or private equity, this communication may be intended as an offer or solicitation for the purchase or sale of such fund(s). All market prices, data and other information are not warranted as to completeness or accuracy and are subject to change without notice. Any comments or statements made herein do not necessarily reflect those of JPMorgan Chase Co., its subsidiaries and affiliates. This transmission may contain information that is privileged, confidential, legally privileged, and/or exempt from disclosure under applicable law. If you are not the intended recipient, you are hereby notified that any disclosure, copying, distribution, or use of the information contained herein (including any reliance thereon) is STRICTLY PROHIBITED. Although this transmission and any attachments are believed to be free of any virus or other defect that might affect any computer system into which it is received and opened, it is the responsibility of the recipient to ensure that it is virus free and no responsibility is accepted by JPMorgan Chase Co., its subsidiaries and affiliates, as applicable, for any loss or damage arising in any way from its use. If you received this transmission in error, please immediately contact the sender and destroy the material in its entirety, whether in electronic or hard copy format. Thank you. Please refer to http://www.jpmorgan.com/pages/disclosures for disclosures relating to UK legal entities. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] dotplot with a log scale
On 10/31/08, glaporta [EMAIL PROTECTED] wrote: Dear all, I would like to create a dotplot (lattice) with a log scale as the function http://www.nabble.com/dotchart-with-log-scale--td5411625.html#a5411625 mydotchart of Johannes Huesing. Is it possible? The usual lattice syntax for this is dotplot(..., scales = list(x = list(log = TRUE))) Does this not work? -Deepayan __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] [ifelse] how to maintain a value from original matrix without probs?
on 10/31/2008 09:59 AM Diogo André Alagador wrote: Dear all, I have a matrix with positive and negative values. From this I would like to produce 2 matrices: 1st - retaining positives and putting NA in other positions 2nd - retaining negatives and putting NA in other positions and then apply rowMeans for both. I am trying to use the function ifelse in the exemplified form: ifelse(A0,A,NA) but by putting A as a 2nd parameter it changes dimensions of the original object. I wonder if I can do this, as it seems not to difficult. thanks in advance A couple of approaches, depending upon the size of the matrix. The first, if the matrix is small-ish: set.seed(1) mat - matrix(sample(-10:10, 100, replace = TRUE), ncol = 10) mat [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10] [1,] -5 -690709 -3 -1-5 [2,] -3 -7 -6238 -474-9 [3,]24306 -1 -1 -3 -2 3 [4,]9 -2 -8 -71 -5 -4 -3 -4 8 [5,] -66 -571 -9305 6 [6,]80 -246 -8 -58 -6 6 [7,]95 -106 -10 -4084-1 [8,]3 10 -2 -8006 -2 -8-2 [9,]3 -38553 -96 -5 7 [10,] -96 -3 -24 -28 10 -7 2 apply(mat, 1, function(x) mean(x[x 0], na.rm = TRUE)) [1] 8.33 4.80 3.60 6.00 4.67 6.40 6.40 [8] 6.33 5.285714 6.00 apply(mat, 1, function(x) mean(x[x 0], na.rm = TRUE)) [1] -4.00 -5.80 -1.75 -4.714286 -6.67 -5.25 [7] -6.25 -4.40 -5.67 -4.60 This way, you avoid splitting the matrix. You did not specify how you might want 0's to be handled, so adjust the logic above accordingly. If the matrix is large, such that that splitting it and using rowMeans() would be faster: mat.pos - mat is.na(mat.pos) - mat = 0 mat.pos [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10] [1,] NA NA9 NA7 NA9 NA NANA [2,] NA NA NA238 NA74NA [3,]243 NA6 NA NA NA NA 3 [4,]9 NA NA NA1 NA NA NA NA 8 [5,] NA6 NA71 NA3 NA5 6 [6,]8 NA NA46 NA NA8 NA 6 [7,]95 NA6 NA NA NA84NA [8,]3 10 NA NA NA NA6 NA NANA [9,]3 NA8553 NA6 NA 7 [10,] NA6 NA NA4 NA8 10 NA 2 rowMeans(mat.pos, na.rm = TRUE) [1] 8.33 4.80 3.60 6.00 4.67 6.40 6.40 [8] 6.33 5.285714 6.00 mat.neg - mat is.na(mat.neg) - mat = 0 mat.neg [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10] [1,] -5 -6 NA NA NA NA NA -3 -1-5 [2,] -3 -7 -6 NA NA NA -4 NA NA-9 [3,] NA NA NA NA NA -1 -1 -3 -2NA [4,] NA -2 -8 -7 NA -5 -4 -3 -4NA [5,] -6 NA -5 NA NA -9 NA NA NANA [6,] NA NA -2 NA NA -8 -5 NA -6NA [7,] NA NA -10 NA -10 -4 NA NA NA-1 [8,] NA NA -2 -8 NA NA NA -2 -8-2 [9,] NA -3 NA NA NA NA -9 NA -5NA [10,] -9 NA -3 -2 NA -2 NA NA -7NA rowMeans(mat.neg, na.rm = TRUE) [1] -4.00 -5.80 -1.75 -4.714286 -6.67 -5.25 [7] -6.25 -4.40 -5.67 -4.60 See ?is.na and note the assignment variant. HTH, Marc Schwartz __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] dot colour in dotchart2 function (Hmisc-package)
Hi, I'm using the dotchart2-function from the Hmisc-package to plot multiple dots on horizontal lines using the add=TRUE option. My question is very simple: How can I change the color of the dots? I couldn't find the right option in the help-pages of the dotchart2-function or the plot.default-function and I tried all obvious possibilities (col, color, dotcol, etc.). Thanks in advance, Rainer WikipediaWictionaryChambers (UK)Google imagesGoogle defineThe Free DictionaryJoin exampleWordNetGoogleUrban DictionaryAnswers.comrhymezone.comMerriam-Webster0 wvcidfjoguarm __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] smooth function advice
RSiteSearch(ewma,rest=func) would have told you there already exists an ewma implementation in the qcc package. -- Bert Gunter -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED] Sent: Friday, October 31, 2008 10:04 AM To: [EMAIL PROTECTED] Cc: r-help@r-project.org Subject: Re: [R] smooth function advice Thanks to Mark Leeds for an implementation of exponentially weighted moving averages as follows which solves this problem. This also resmooths recursively. ewma-function(x,lambda = .5, init = (1-lambda)*.raw[good.ind][1],order=1) { .raw - unclass(coredata(x)) good.ind - !is.na(.raw) # determine good values if(order1) ewma(ewma(x,lambda,init,order=1),lambda,init,order=order-1) else { # work with 'non-zoo' data for speed and then recombine .raw[good.ind] - filter(lambda * .raw[good.ind], filter=(1-lambda), method='recursive',init=coredata(init)) zoo(.raw, index(x)) # create zoo object for return } } Tolga I Uzuner/JPMCHASE 30/10/2008 15:36 To r-help@r-project.org cc Subject smooth function advice Dear R Users, I am looking for a smoothing function with the following characteristics for a time series of data: - at each date, should only use data up to that date (so, right aligned and not centered) - should return a smoothed series of length equal to the original time series: - for a one-day time series, just returns that day - this means the front part of the series will not be as smooth... that's ok - if the original time series has length 10, the returned smoothed time series has length 10 - there should be some control over smoothness Any suggestions for a specific approach/package in R ? Thanks, Tolga Generally, this communication is for informational purposes only and it is not intended as an offer or solicitation for the purchase or sale of any financial instrument or as an official confirmation of any transaction. In the event you are receiving the offering materials attached below related to your interest in hedge funds or private equity, this communication may be intended as an offer or solicitation for the purchase or sale of such fund(s). All market prices, data and other information are not warranted as to completeness or accuracy and are subject to change without notice. Any comments or statements made herein do not necessarily reflect those of JPMorgan Chase Co., its subsidiaries and affiliates. This transmission may contain information that is privileged, confidential, legally privileged, and/or exempt from disclosure under applicable law. If you are not the intended recipient, you are hereby notified that any disclosure, copying, distribution, or use of the information contained herein (including any reliance thereon) is STRICTLY PROHIBITED. Although this transmission and any attachments are believed to be free of any virus or other defect that might affect any computer system into which it is received and opened, it is the responsibility of the recipient to ensure that it is virus free and no responsibility is accepted by JPMorgan Chase Co., its subsidiaries and affiliates, as applicable, for any loss or damage arising in any way from its use. If you received this transmission in error, please immediately contact the sender and destroy the material in its entirety, whether in electronic or hard copy format. Thank you. Please refer to http://www.jpmorgan.com/pages/disclosures for disclosures relating to UK legal entities. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Digest setting won't take
I've been trying to set my email setting to Digest Mode for a while, but it doesn't seem to take. I log through to the Unsubscribe or edit options button on the Web interface to the R-help mailing list page. I scroll down to the Set Digest Mode box and change the radio button to On. I click the Submit My Changes button which takes me back to the mailing list log-in page. I log back in and the setting is Off. It looks like all settings are re-set to the default. How do I change these settings? Thanks Robert Farley Metro 1 Gateway Plaza Mail Stop 99-23-7 Los Angeles, CA 90012-2952 Voice: (213)922-2532 Fax:(213)922-2868 www.Metro.net [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] help with contrasts for a binomial 3-way GLM
Hi I am a new user the R and I am very grateful for all your help but... I have a problem and I can't resolve yet. I am trying to get the contrasts for a binomial 3-way GLM (T= 4 temperature, t= 2 time and c= 2 substrate levels, plus treatment control) in total they are 17 treatments. I have tried with the glht but this function only work for 1-way GLM, acacia-cbind(g,N-g) acacia1-glm(acacia~tratamiento,binomial) acacia2-glht(acacia1, linfct = mcp(tratamiento = Tukey)) summary(acacia2) Simultaneous Tests for General Linear Hypotheses Multiple Comparisons of Means: Tukey Contrasts Fit: glm(formula = acacia ~ tratamiento, family = binomial) Linear Hypotheses: Estimate Std. Error z value p value B - A == 02.9073 1.0395 2.797 0.2417 C - A == 04.0860 1.0227 3.995 0.01 ** D - A == 04.8458 1.0207 4.748 0.01 *** E - A == 03.0445 1.0362 2.938 0.1727 F - A == 03.3399 1.0306 3.241 0.0744 .etc. If the GLM is with interactions don't work Ø acacia3-glm(acacia~T*t*c,binomial) Øglht(acacia3, linfct = mcp(T = Tukey)) Ø Error in modelparm.default(model, ...) : Ø dimensions of coefficients and covariance matrix don't match Ø Warning message: Ø In mcp2matrix(model, linfct = linfct) : Ø covariate interactions found -- default contrast might be inappropriate Other function could be simtes but this was deprecated from package multcomp There are others options to get comparisons contrasts for model GLM with interactions? or this test don't exist ? I hope help me [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Quantile Regression for Longitudinal Data:error message
If you are going to insist on doing such things you will have to learn to read the documentation. In this case if you do a traceback() you will see that the error is occurring in rq.fit.slm and when you do ?rq.fit.slm you will see that there are several storage sizes that can be adjusted: nsubmax: upper bound of the dimension of lindx in Cholesky factorization; computed automatically inside the routine if missing. tmpmax: upper bound of the working array in Cholesky factorization; computed automatically inside the routine if missing. nnzlmax: upper bound of the non-zero entries in the Cholesky factor L; computed automatically inside the routine if missing. cachsz: size of the cache on the machine; default to 64. Since you don't give, as stipulated by the posting guide, a reproducible example, one can only speculate about what you are trying to do, but it appears that you are trying to simultaneously estimate 99 quantiles with a sample size of length(w) = 99 observations and this is Unsound. I would also like to repeat my earlier comment that this is NOT a proper R help question since it refers not to something that is in base R, nor even something in a package, but to code that I happen to have posted to complement a paper that was published several years ago. url:www.econ.uiuc.edu/~rogerRoger Koenker email[EMAIL PROTECTED]Department of Economics vox: 217-333-4558University of Illinois fax: 217-244-6678Champaign, IL 61820 On Oct 31, 2008, at 11:32 AM, Helen Chen wrote: Quantile Regression for Longitudinal Data. Hi, I am trying to estimate a quantile regression using panel data. I am trying to use the model that is described in Dr. Koenker's article. So I use the code the that is posted in the following link: http://www.econ.uiuc.edu/~roger/research/panel/rq.fit.panel.R I am trying to change the number quantiles being estimated. I change the codes about w and taus ,as following w=c(.01,.02,.03,.04,.05,.06,.07,.08,.09,.1,.11,.12,.13,.14,.15,.16,.17 ,.18,.19,.2,.21,.22,.23,.24,.25,.26,.27,.28,.29,.3,.31,.32,.33,.34,.35 ,.36,.37,.38,.39,.4,.41,.42,.43,.44,.45,.46,.47,.48,.49,.5,.49,.48,.47 ,.46,.45,.44,.43,.42,.41,.4,.39,.38,.37,.36,.35,.34,.33,.32,.31,.3,.29 ,. 28,.27,.26,.25,.24,.23,.22,.21,.2,.19,.18,.17,.16,.15,.14,.13,.12,.11 ,.1,.09,.08,.07,.06,.05,.04,.03,.02,.01) ,taus=(1:99)/100,lambda = 1 But I get error message: .local(x, pivot, ...) : Increase tmpmax So I am wondering if I am doing something wrong or I mistake w's meaning. Thanks I really would appreciate some suggestions. Best Helen Chen -- View this message in context: http://www.nabble.com/Quantile-Regression-for-Longitudinal-Data%3Aerror-message-tp20269386p20269386.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] reshape bug?
Hadley et al., I was using the cast function to reshape some data (aggregate a melted data frame) and I did not put in the fill and for the most part the values that came out were fine, but there were value great than an order of magnitude from the actual value. When I put in the fill argument everything is okay. I don't provide a reproducible example because the data set is to large to post to the list, but if you would like to see it then I can provide it. #this works cast(x, Date+RiverMile+location~Order, sum, fill=0) #this doesn't cast(x, Date+RiverMile+location~Order, sum) R version 2.8.0 (2008-10-20) i386-pc-mingw32 locale: LC_COLLATE=English_United States.1252;LC_CTYPE=English_United States.1252;LC_MONETARY=English_United States.1252;LC_NUMERIC=C;LC_TIME=English_United States.1252 attached base packages: [1] splines grid stats graphics grDevices utils datasets methods base other attached packages: [1] ggplot2_0.7MASS_7.2-44RColorBrewer_1.0-2 proto_0.3-8reshape_0.8.1 plyr_0.1.1 thanks -- Stephen Sefick Research Scientist Southeastern Natural Sciences Academy Let's not spend our time and resources thinking about things that are so little or so large that all they really do for us is puff us up and make us feel like gods. We are mammals, and have not exhausted the annoying little problems of being mammals. -K. Mullis __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Compiling R Packages
I'm currently using R version 2.7.2. Not the newest, but not even six months old either. Thanks for the idea. -Original Message- From: Erin Hodgess [mailto:[EMAIL PROTECTED] Sent: Thursday, October 30, 2008 4:06 PM To: Cahill, Joshua Cc: r-help@r-project.org Subject: Re: [R] Compiling R Packages Which version of R are you using, please? I had that problem with some earlier versions of R. On Thu, Oct 30, 2008 at 3:59 PM, Cahill, Joshua [EMAIL PROTECTED] wrote: I am working on a SLES 10 cluster with R available on it. To better use my resources, I want to use Rmpi, but I am having a difficult time installing it. I have set the $R_LIBS variable correctly. The MPI libraries are available through the PGI compiler, but this R instance was compiled with GCC. The Rmpi package can be compiled and installed with %R CMD INSTALL Rmpi_0.5-5.tar.gz --configure-args=CC=pgcc --with-mpi=/shared/pgi725/linux86-64/7.2/mpi/mpich But when entering R and loading the library, I get %R library(Rmpi) Error in dyn.load(file, DLLpath = DLLpath, ...) : unable to load shared library '/home/username/.Rlib/Rmpi/libs/Rmpi.so': /home/username/.Rlib/Rmpi/libs/Rmpi.so: undefined symbol: __pgdbg_stub Error in library(Rmpi) : .First.lib failed for 'Rmpi' Error in dyn.unload(file.path(libpath, libs, paste(Rmpi, .Platform$dynlib.ext, : dynamic/shared library '/home/username/.Rlib/Rmpi/libs/Rmpi.so' was not loaded The undefined symbol is for the PGI debugger. It doesn't matter if the $CC variable is gcc, pgcc, or mpicc, and all three do show up in the make output during the configure process. Thanks for your help, Joshua __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Erin Hodgess Associate Professor Department of Computer and Mathematical Sciences University of Houston - Downtown mailto: [EMAIL PROTECTED] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] R-WinEdt unexcpected Asian characters
Saskia Sandring wrote: Dear list! I have encountered an odd problem with R-WinEdt. The program is mostly running fine and I had no problems with the installation. But several times a day I'm typing in R-WinEdt there appears a menu with Chinese characters where the curser is and I'm not able to type normally. I don’t know any other way to stop it except restarting R-WinEdt. After that it works normally for a while until it happens again. I was not able to figure out when it happens. Sometimes after error messages in R, but not always and I am never aware of having typed a certain key combination or anything like that. Is there anything I can change in the settings to prevent this from happening? I'm using R 2.8.0. and WinEdt 5.5. in Windows XP. My regional options in Windows are set to Sweden and Swedish. With earlier versions of R and WinEdt I did not have this problem. However, this is a different computer. It should not be related to R (no direct link between both programs), and probably not to RWinEdt (because I have not had a second report like this). Best wishes, Uwe Ligges Thank you for your help! Saskia __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Digest setting won't take
On Friday, 31 October 2008, 10:46 (UTC-0700), Farley, Robert wrote: I've been trying to set my email setting to Digest Mode for a while, but it doesn't seem to take. I think I had such a problem at some point, and I think it was because I didn't have cookies enabled in my browser. m. -- Marianne Promberger Graduate student in Psychology http://www.psych.upenn.edu/~mpromber __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] subset with variable number of subset arguments
Thank you very much! Brar Erik Iverson schrieb: I think subset(set, InfCode %in% x) does it, where 'x' is your vector of ids (e.g., c(1,3,9) in your example at the bottom) Brar Piening wrote: I've got a big data.frame from which I need to extract data based on a variable number of Id's (InfCode). Until now I've been using the following dull solution as I never needed to search for more than 5 codes. Now my needs have increased faster than my R skills did and I need to call my function with about 25 values for x. There has to be a *apply or even simpler solution which (after RTM) I'm not able to figure out. Can anybody direct me to a possible solution? Regards, Brar GetInfCodeSubset - function(set, x){ if(1 == length(x)) return(subset(set, (set$InfCode == x[1]))) else if(2 == length(x)) return(subset(set, (set$InfCode == x[1]) | (set$InfCode == x[2]))) else if(3 == length(x)) return(subset(set, (set$InfCode == x[1]) | (set$InfCode == x[2]) | (set$InfCode == x[3]))) else if(4 == length(x)) return(subset(set, (set$InfCode == x[1]) | (set$InfCode == x[2]) | (set$InfCode == x[3]) | (set$InfCode == x[4]))) else if(5 == length(x)) return(subset(set, (set$InfCode == x[1]) | (set$InfCode == x[2]) | (set$InfCode == x[3]) | (set$InfCode == x[4]) | (set$InfCode == x[5]))) else if(6 == length(x)) return(subset(set, (set$InfCode == x[1]) | (set$InfCode == x[2]) | (set$InfCode == x[3]) | (set$InfCode == x[4]) | (set$InfCode == x[5]) | (set$InfCode == x[6]))) else stop(Too many elements in x) } set - data.frame(PatId = c(1:100), InfCode = rep(c(1:20), 5)) GetInfCodeSubset(set, c(1,3,9)) -- __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] replace() error: new columns would leave holes after existing columns
Hello, I have a problem with using replace() to convert a vector of dates from -mm-dd to julian date. For example, I type replace(x,2004-05-14,134) and I receive an error: Error in `[-.data.frame`(`*tmp*`, list, value = 134) : new columns would leave holes after existing columns If I can successfully convert, I have a script that will convert all of the dates in the vector (i.e., the same line repeated with the next -mm-dd and corresponding julian day). Any help would be much appreciated. Thank you, Kirsten Kirsten Barrett Mendenhall Postdoctoral Fellow, USGS Alaska Science Center 4210 University Drive Anchorage, AK [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] AIC for quasipoisson link
on 10/31/2008 01:07 PM [EMAIL PROTECTED] wrote: Dear fellows, I'm trying to extract the AIC statistic from a GLM model with quasipoisson link. The formula I'm referring to is AIC = -2(maximum loglik) + 2df * phi with phi the overdispersion parameter, as reported in: Peng et al., Model choice in time series studies os air pollution and mortality. J R Stat Soc A, 2006; 162: pag 190. Unfortunately, the function logLik doesn't work for a quasipoisson link. Do you know a fast method to extract the AIC for these models? Thanks in advance I was under the impression that there is no log likelihood for quasi* family models, thus no AIC, which is why they are not calculated/printed in the glm() summary outputs. If you want to model overdispersed data and need the AIC, you should look at glm.nb() in MASS for a negative binomial model: library(MASS) ?glm.nb This would also avail you of the anova.glm() methods for comparing models, which the quasi* families would not. You might also want to look at: http://cran.r-project.org/web/packages/pscl/vignettes/countreg.pdf which is the vignette from the pscl package. HTH, Marc Schwartz __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] how to compute a roc curve
There are various packages that do ROC curves (others have mentioned some, you can search for the others), but ROC curves are not that hard to do by hand in R. Using the classic iris dataset, we can use Sepal.Length as the test/score and species=='virginica' as the true pos/neg that we are trying to predict. One way to compute the ROC curve, auc and values is: xx - c( -Inf, sort(unique(iris$Sepal.Length)), Inf ) sens - sapply(xx, function(x) with(iris, + mean( Sepal.Length[Species=='virginica'] = x ) ) ) spec - sapply(xx, function(x) with(iris, + mean( Sepal.Length[Species!='virginica'] x ) ) ) plot( 1-spec, sens, type='l' ) ch - chull(1-spec,sens) lines(1-spec[ch], sens[ch], col='green') integrate( approxfun(spec, sens), 0, 1) # Area under curve Hope this helps, -- Gregory (Greg) L. Snow Ph.D. Statistical Data Center Intermountain Healthcare [EMAIL PROTECTED] 801.408.8111 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] project.org] On Behalf Of Pau Marc Munoz Torres Sent: Friday, October 31, 2008 8:25 AM To: r-help@r-project.org Subject: [R] how to compute a roc curve Hi, I'm trying to set up a prediction software, now i testing the performance of my method, so i need to calculate a ROC curve, specially auc, cut-off, sens and spec, i just looking at ROCH package, but it's a mass for me, i'm not a math guy and I'm getting lost Could any of you recommend me an easy-to-use package to do this task? i just have a list of positive/negative samples and his score on my program. can I compute a roc curve with this? thanks pau -- Pau Marc Muñoz Torres Laboratori de Biologia Computacional Institut de Biotecnologia i Biomedicina Vicent Villar Universitat Autonoma de Barcelona E-08193 Bellaterra (Barcelona) telèfon: 93 5812807 Email : [EMAIL PROTECTED] [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] dotplot with a log scale
Perfectly... thanx Deepayan Deepayan Sarkar wrote: On 10/31/08, glaporta [EMAIL PROTECTED] wrote: Dear all, I would like to create a dotplot (lattice) with a log scale as the function http://www.nabble.com/dotchart-with-log-scale--td5411625.html#a5411625 mydotchart of Johannes Huesing. Is it possible? The usual lattice syntax for this is dotplot(..., scales = list(x = list(log = TRUE))) Does this not work? -Deepayan __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- View this message in context: http://www.nabble.com/dotplot-with-a-log-scale-tp20265718p20273083.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Formula for Xi
Am 30.10.2008 um 16:53 schrieb [EMAIL PROTECTED]: Hallo! ?xml:namespace prefix = o ns = urn:schemas-microsoft-com:office:office /o:p/o:p [...] [[alternative HTML version deleted]] This is a rare case where I would have liked to see the HTML version in order to see how these o:p tags are interpreted (paragraphs? tab advances?). But then I doubt if MS-Office generated HTML will be any more conclusive. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] AIC for quasipoisson link
Marc Schwartz marc_schwartz at comcast.net writes: on 10/31/2008 01:07 PM Antonio.Gasparrini at lshtm.ac.uk wrote: I'm trying to extract the AIC statistic from a GLM model with quasipoisson link. The formula I'm referring to is AIC = -2(maximum loglik) + 2df * phi with phi the overdispersion parameter, as reported in: Peng et al., Model choice in time series studies os air pollution and mortality. J R Stat Soc A, 2006; 162: pag 190. I was under the impression that there is no log likelihood for quasi* family models, thus no AIC, which is why they are not calculated/printed in the glm() summary outputs. Yes, but ... this is a matter of some disagreement. Long answer: The purist position (hi Prof. Ripley) is that quasi-likelihood estimation does not produce a likelihood and should not return one. A common position in applied statistics (I think starting with a paper by Lebreton, but I can't find the ref right now: see refs below) is that dividing the log-likelihood of a regular likelihood fit by the estimated scale (overdispersion) parameter of the quasi- variant gives a quasilikelihood that can be used to compute a quasi-AIC that can then be used in model selection. Short answer: I think that if you fit the non-quasi version of the model (ie. Poisson family in your case) and extract the likelihood from it, then divide by the overdispersion parameter estimated from the quasi variant, that should give you what you want. By the way, the formula quoted above looks funny. Shouldn't it be QAIC = -2(maximum loglik)/phi + 2df ? The formula quoted above (phi times my version) should give the same ordering, but model weights and interpretations of QAIC differences will be wrong. cheers Ben Bolker Anderson, D. R., K. P. Burnham, and G. C. White. 1994. AIC model selection in overdispersed capture-recapture data. Ecology 75, no. 6: 1780-1793. Richards, Shane A. 2008. Dealing with overdispersed count data in applied ecology. Journal of Applied Ecology 45: 218-227. doi:10./j.1365-2664.2007.01377.x. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] why does sample(x, n) give the same n items in every separate runs?
Hello, I'd like to think all respondents for your advises and suggestions. I am sorry for not including the ordered gene expression data frame (largedf) since it is large. The problem is solved now by Greg Down by including the command: rm(.Random.seed) in the code, perhaps I used some R packges to process that expression data which set the seed on the Random Number Generator to a fixed value. Thanks again, Joshua - Original Message - From: Charles C. Berry [EMAIL PROTECTED] To: Shuhua Zhan [EMAIL PROTECTED] Cc: r-help@r-project.org Sent: Thursday, October 30, 2008 9:07:09 PM GMT -05:00 US/Canada Eastern Subject: Re: [R] why does sample(x, n) give the same n items in every separate runs? Run help.request(). In particular you need to attend to this part: Have you written example code that is - minimal - reproducible - self-contained - commented using data that is either - constructed by the code - loaded by data() - reproduced using dump(mydata, file = ) have you checked this code in a fresh R session (invoking R with the --vanilla option if possible) and is this code copied to the clipboard? (y/n) Once you get this far and can honestly type 'y', if you have not found your error, you are ready to post a query. HTH, Chuck On Thu, 30 Oct 2008, Shuhua Zhan wrote: Hello R users, I have gene expression data of two groups of genes (large and small). Gene expression intensities of those genes are classified into 1 to 10 levels. What I want is to make a random set of genes that have the same levels as the small group from large group using sample(). I used smallvec to hold the number of genes in each levels (1 to 10) for small group, largevec for large group. I ordered the gene expression data frame of large group (largedf) by the levels and randomly chose the genes with same level as small group. Using the code below I can get the random set of genes from lagre group with same levels for small group. But I got the same set of genes when I run the code in every other runs in Linux. This gives me a doubt in my result. smallvec-c(8,12,9,6,13,20,16,11,8,5) ## the No. of genes in levels 1 to 10 largevec-c(400,300,550,600,210,420,380,600,450,500) generdm-c() ## a random set of genes for( i in 1:length(smallvec)){ generdm-c(generdm,sample(rownames(largedf)[sum(largevec[0:(i-1)],1):sum(largevec[0:i])],smallvec[i])) ## rownames(largedf) gives gene names ordered by levels in large group } Could you please help me out? Thanks a lot!! Josh __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Charles C. Berry(858) 534-2098 Dept of Family/Preventive Medicine E mailto:[EMAIL PROTECTED] UC San Diego http://famprevmed.ucsd.edu/faculty/cberry/ La Jolla, San Diego 92093-0901 __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] AIC for quasipoisson link
Ben, I think the reference you're searching for is the one below @ARTICLE{Lebreton1992, author = {Lebreton, J.-D. and Burnham, K. P. and Clobert, J. and Anderson, D. R.}, title = {Modeling survival and testing biological hypotheses using marked animals: a unified approach with case studies}, journal = {Ecological {M}onographs}, year = {1992}, volume = {62}, pages = {67-118}, keywords = {Modeling, survival, Capture-recapture}, pdf = {Lebreton1992.pdf} } Cheers, C Selon Ben Bolker [EMAIL PROTECTED]: Marc Schwartz marc_schwartz at comcast.net writes: on 10/31/2008 01:07 PM Antonio.Gasparrini at lshtm.ac.uk wrote: I'm trying to extract the AIC statistic from a GLM model with quasipoisson link. The formula I'm referring to is AIC = -2(maximum loglik) + 2df * phi with phi the overdispersion parameter, as reported in: Peng et al., Model choice in time series studies os air pollution and mortality. J R Stat Soc A, 2006; 162: pag 190. I was under the impression that there is no log likelihood for quasi* family models, thus no AIC, which is why they are not calculated/printed in the glm() summary outputs. Yes, but ... this is a matter of some disagreement. Long answer: The purist position (hi Prof. Ripley) is that quasi-likelihood estimation does not produce a likelihood and should not return one. A common position in applied statistics (I think starting with a paper by Lebreton, but I can't find the ref right now: see refs below) is that dividing the log-likelihood of a regular likelihood fit by the estimated scale (overdispersion) parameter of the quasi- variant gives a quasilikelihood that can be used to compute a quasi-AIC that can then be used in model selection. Short answer: I think that if you fit the non-quasi version of the model (ie. Poisson family in your case) and extract the likelihood from it, then divide by the overdispersion parameter estimated from the quasi variant, that should give you what you want. By the way, the formula quoted above looks funny. Shouldn't it be QAIC = -2(maximum loglik)/phi + 2df ? The formula quoted above (phi times my version) should give the same ordering, but model weights and interpretations of QAIC differences will be wrong. cheers Ben Bolker Anderson, D. R., K. P. Burnham, and G. C. White. 1994. AIC model selection in overdispersed capture-recapture data. Ecology 75, no. 6: 1780-1793. Richards, Shane A. 2008. Dealing with overdispersed count data in applied ecology. Journal of Applied Ecology 45: 218-227. doi:10./j.1365-2664.2007.01377.x. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Is there a way to vectorize this? [with correction]
** Sorry to repost. I forgot to include a function necessary to make the example work ** I apologize up front for this being a little long. I hope it's understandable. Please let me know if I need to clarify anything. Several months ago I wrote a series of functions to help me take my R analyses and build custom reports in html files. Each function either builds or modifies a string of html code that can then be written to a file to produce the desired output. To make modifications in the html code, I've placed 'markers' around certain characteristics that I might want to change. For instance, the alignment characteristics have an 'algnmark' on either side of them. When I wish to change the alignment, I can find where these markers are, determine their location, and replace the contents between them. I've been using the functions for a few months now, and am pleased with the utility. Unfortunately, as I was writing these, I wasn't very strong with my vectorization skills and relied on for loops (lots of for loops) to get through the work. So while I'm pleased with the utility, I've been trying to optimize the functions by vectorizing the for loops. At this point, I've hit a small snag. I have a situation where I can't seem to figure out how to vectorize the loop. Part of me wonders if it is even possible. The scenario is this: I run a string of code through the loop, on each pass, the section of code in need of modification is identified and the changes are made. When this is done, however, the length of the string changes. The change in length needs to be recognized in the next pass through the loop. Okay, some code to illustrate what I mean. This first function formats the html file. I only include it because it will be necessary to create illustrate what the function is doing. I am eliminating all comments and spacing from the code for brevity. #*** Start of html.file.start 'html.file.start' - function(title, size=11, font=Times New Roman){ size - format(floor(size),nsmall=1) code - paste( html xmlns:o='urn:schemas-microsoft-com:office:office\' xmlns:w=\'urn:schemas-microsoft-com:office:word\' xmlns=\'http://www.w3.org/TR/REC-html40\' head meta http-equiv=Content-Type content=\'text/html; charset=windows-1252\' meta name=ProgId content=Word.Document meta name=Generator content=\'Microsoft Word 11\' meta name=Originator content=\'Microsoft Word 11\' style !-- /* Style Definitions */ p.MsoNormal, li.MsoNormal, div.MsoNormal p.MsoEndnoteText, li.MsoEndnoteText, div.MsoEndnoteText {margin-top:2.0pt; margin-right:0in; margin-bottom:0in; margin-left:.15in; margin-bottom:.0001pt; text-indent:-.15in; mso-pagination:none; font-size:9.0pt; mso-bidi-font-size:10.0pt; font-family:'Times New Roman'; mso-fareast-font-family:'Times New Roman';} p.Textbody, li.Textbody, div.Textbody-- /style , title,title,/title /head body lang=EN-US style=\'tab-interval:.5in;, textmark; font-size:,size,pt; textmark;, fontmark; font-family:,font,; fontmark;\', sep=) return(code) } # End of html.file.start # Start of html.text 'html.text' - function(text, size=11, font=Times New Roman, align=left, title){ size - format(floor(size),nsmall=1) if(missing(title)) title - else title - paste(br/,title) title - paste(b,title,/bbr/\n,sep=) code - paste( p class=MsoNormal , algnmark align=,align, algnmark span class=GramE style=\', textmark; font-size:,size,pt; textmark;, fontmark; font-family:,font,; fontmark;, stylemark; font-weight:normal; font-style:normal;, text-decoration:none; stylemark;\', title,text, /span /p,sep=) return(code) } #** End of html.text So here is the function I'm trying to vectorize. #*** Start of html.align html.align - function(code,new.align=left){ #* Create a string to replace the current alignment setting. align - paste( align=,new.align, ,sep=) #* Function to pass to sapply. This is handy when 'code' #* is a vector. f1 - function(code,align=align){ mark - unlist(gregexpr(algnmark,code)) #* Get positions of markers if(mark[1]0){ odd - seq(1,length(mark),by=2) #* odd elements are starting marker evn - seq(2,length(mark),by=2) #* even elements are ending marker mark[odd] - mark[odd]+9 #* These two lines determine the starting mark[evn] - mark[evn]-1 #* and ending elements of the substring to #* be replaced for(i in 1:length(odd)){ l.old - nchar(code) #* store the length of the code segment. old.align - substr(code,mark[odd[i]],mark[evn[i]])
[R] ctree to rules conversion function
Hi, Similar to the R function of converting rpart into rules (see example: http://books.huihoo.org/data-mining-desktop-survival-guide/Convert_Tree. html_ is there any open code to convert ctree (from party) into rule format? This would be very useful. thanks Dhruv Dhruv Sharma Credit Risk Policy and Modeling Manager Pentagon Federal Credit Union 703-838-1086 [EMAIL PROTECTED] mailto:[EMAIL PROTECTED] [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] how to compute a roc curve
Pau Marc Munoz Torres wrote: Hi, I'm trying to set up a prediction software, now i testing the performance of my method, so i need to calculate a ROC curve, specially auc, cut-off, sens and spec, i just looking at ROCH package, but it's a mass for me, i'm not a math guy and I'm getting lost The use of any cutoffs will result in inefficient and arbitrary analyses, and note that sensitivity and specificity are improper scoring rules (they are optimized by a bogus model). Frank Could any of you recommend me an easy-to-use package to do this task? i just have a list of positive/negative samples and his score on my program. can I compute a roc curve with this? thanks pau __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Frank E Harrell Jr Professor and Chair School of Medicine Department of Biostatistics Vanderbilt University __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] PAM: how to get the best number of clusters
Maura, No, in general it is not possible. Depending on your goodness-of-clustering metric (and there are many besides silhouette width), the results may demonstrate multiple peaks and be severely non-monotonic. Perhaps more problematic, silhouette width is especially difficult. It is local, as opposed to global, so that it only compares an object to the cluster to which it is assigned and to the most similar cluster. Each object in a cluster may be most similar to a different cluster, so that re-assigning a single plot from one cluster to another requires recalculating the whole mess for all clusters. In addition, despite the utility of a mean silhouette width, it is insufficient as a metric in that many different solutions might give the same mean silhouette width and yet be very different in numbers of reversals (negative silhouette widths) or variance in within-cluster silhouette width. The beauty of the silhouette width concept is the plot, and that means you have to actually look at the plot of the solution, and not simply accept a mean silhouette width. I think you need to consider realistically how many clusters you want, and limit yor consideration to solutions of approximately that number. Good luck, Dave R. - - David W. Roberts office 406-994-4548 Professor and Head FAX 406-994-3190 Department of Ecology email [EMAIL PROTECTED] Montana State University Bozeman, MT 59717-3460 Maura E Monville wrote: My problem is that I have already a distance (similarity) matrix generated outside R through a C++ code because the criteria to calculate the distance between pairs of objects are none of the standard criteria implemented in R. If I got it right, but I might be mistaken, stepFlexclust() performs the clusters layout optimization by calling either one of kcca or cclust which calculate their own similarity matrix. I just need a function or method to optimize the clusters number no matter how the similarity matrix has been generated and no matter which clustering function I use (PAM). Is this at all possible ? Thank you very much, Maura On Fri, Oct 31, 2008 at 12:18 AM, Dylan Beaudette [EMAIL PROTECTED] wrote: On Thursday 30 October 2008, Maura E Monville wrote: I have the book you mentioned. It basically describes the silhouette method. I do not have it handy as I moved so it is still in some box. However I cannot remember that book providing any other criterion to find the best clusters number. On the other hand I have the same problem with hierarchical clustering techniques. I use clusters as exploratory analysis because I do not have any a-priori knowledge that helps me make a choice. How can multivariate analysis help? I launched a loop where the silhouette test follows PAM which is passed a clusters number increased by 1 at each iteration. Since I am observing that the silhouette value is now oscillating among negative numbers, I wonder whether I can assume that it can only grow worse once it has turned negative the first time so leave the loop after the first negative number and choose the clusters number associated with the biggest positive silhouette value. This procedure would spare a lot of CPU time. Another approach might involve the stepFlexclust() from the flexclust package. See the manual page for this function for examples. Dylan Thank you very much, Maura On Thu, Oct 30, 2008 at 7:25 PM, Dylan Beaudette [EMAIL PROTECTED]wrote: On Thursday 30 October 2008, Maura E Monville wrote: I have a pretty big similarity matrix (2870x2870). I will produce even bigger ones soon. I am using PAM to generate clusters. The desired number of output clusters is a PAM input parameter. I do not know a-priopri what is the best clusters layout . I resorted to the silhouette test. It takes forever as I have to run PAM with all possible numbers of clusters. I wonder whether there is some faster method, either a s/w code or some theoretical guidelines, to get the optimum clusters number. Thank you very much, This is a very general topic in the field of multivariate analysis. There really isn't any way to know the 'correct' number of clusters, however there are several metrics that can give you an indication of how messy your data are. For information on the methods in the cluster package, see this book: Kaufman, L. Rousseeuw, P. J. Finding Groups in Data An Introduction to Cluster Analysis Wiley-Interscience, 2005 Otherwise, consider a book on multivariate analysis. Alternatively, try a hierarchical clustering approach, and look for meaningful groupings. Some thing like this: d - diana(daisy(your_data_matrix)) d.hc - as.hclust(d) d.hc$labels - your_data_matrix$id plot(d.hc) Cheers, Dylan -- Dylan Beaudette Soil Resource Laboratory
Re: [R] Digest setting won't take
No Joy. :-( I've tried with both IE (7.0) and Firefox (3.0.3) and the same thing happens. Firefox has cookies enabled, and I set IE to explicitly allow http://www.r-project.org/ cookies. Robert Farley Metro www.Metro.net -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Marianne Promberger Sent: Friday, October 31, 2008 12:17 To: r-help@r-project.org Subject: Re: [R] Digest setting won't take On Friday, 31 October 2008, 10:46 (UTC-0700), Farley, Robert wrote: I've been trying to set my email setting to Digest Mode for a while, but it doesn't seem to take. I think I had such a problem at some point, and I think it was because I didn't have cookies enabled in my browser. m. -- Marianne Promberger Graduate student in Psychology http://www.psych.upenn.edu/~mpromber __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Why are these different?
I want to import some target values to normalize elements of a dataframe. I'm summarizing the existing totals using as.matrix(xtabs(~tripid_nu+lineon, data=SurveyData)) and reading a .csv table of target values. I can't even get them to list out as similar data items. My idea is to do something like this, but I don't know how to do it in R SurveyData$X1 = NewTargetX1Sums[Trip,Stop] / CurrentX1Sums[Trip,Stop] That is; a different value for X1 for each [Trip,Stop] combination Should I be using as.matrix? Why doesn't NewTargetX1Sums name the rows when CurrentX1Sums does? Why does NewTargetX1Sums have dots in the label spaces? Do I have the right data organization for my problem? What I've done: SurveyData - read.spss(C:/Data/R/orange_delivery.sav, use.value.labels=TRUE, max.value.labels=Inf, to.data.frame=TRUE) NewTargetData - read.table(C:/Data/R/NewTarget.csv, header=TRUE, sep=,, na.strings=NA, dec=.) CurrentX1Sums - as.matrix(xtabs(~tripid_nu+lineon, data=SurveyData)) CurrentX2Sums - apply(CurrentX1Sums, 1, sum) NewTargetX1Sums - as.matrix(NewTargetData) NewTargetX2Sums - apply(NewTargetX1Sums, 1, sum) CurrentX1Sums lineon tripid_nu Warner Center De Soto Pierce College Tampa Reseda Balboa Woodley Sepulveda Van Nuys Woodman Valley College Laurel Canyon North Hollywood 9011880 1 0 2 1 0 2 1 00 0 1 0 0 9011890 0 0 0 0 0 0 1 00 0 0 1 0 9011960 1 1 2 0 1 1 0 13 2 1 0 0 9011970 0 0 0 0 1 0 0 16 1 1 1 14 9012040 1 1 1 3 2 7 1 11 0 0 0 0 {Snip} NewTargetX1Sums tripid_nu Warner.Center De.Soto Pierce.College Tampa Reseda Balboa Woodley Sepulveda Van.Nuys Woodman Valley.College Laurel.Canyon North.Hollywood [1,] 9011880 5 2 2 2 2 2 2 22 2 6 4 1 [2,] 9011890 1 1 1 1 1 1 2 11 1 1 2 1 [3,] 9011960 2 2 2 1 2 2 1 23 2 2 1 1 [4,] 9011970 1 1 1 1 2 1 1 26 2 2 2 24 [5,] 9012040 2 2 2 3 2 7 2 22 1 1 1 1 {Snip} Robert Farley Metro 1 Gateway Plaza Mail Stop 99-23-7 Los Angeles, CA 90012-2952 Voice: (213)922-2532 Fax:(213)922-2868 www.Metro.net [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Estimating R2 value for a unit-slope regression
Dear all, I am in need to estimate the amount of variation explained in a variable after simulations that produce a predictor which is in the same units as the dependent variable (numbers of species). Since the dependent and predictor variables are the same, I would think the most appropriate analysis would be a regression constrained to have an intercept of 0 and a slope of 1. I am trying to write a piece of R code to do this, but I am running into some problems, so I wanted to ask for your advice. I have inverstigated 3 approaches, and i am including a jpg file with the behaivour of these three pieces of code (R2 values as a function of the slope of the ols regression). I also included the regular R2 value from the ols regression for comparison (black symbols in figure). R2 for a regression can be calculated by the formula: R2= (SSY-SSE)/SSY; so: #1 Green symbols in figure SSY-sum((y-mean(y))^2) SSE-SSE-sum((y-x)^2) R2-(SSY-SSE)/SSY where y is the dependent and x the predictor variables respectively, of course. However, I am running into trouble because some times the residual sum of squares (SSE) is larger than the SS of the dependent variable (SSY) and I end up having negative R2s which of course make no sense. Another way to put the same formula is: R2=SSR/SSR+SSE; so: #2 Blue symbols in figure SSR-sum((x-mean(y))^2) SSE-sum((y-x)^2) R2-SSR/(SSR+SSE) This approach behaves beter in the sense that it stays within the 0 to 1 expected range, it peaks when the slope is equal to 1, but its decal as the slope moves away from 1 is too slow, and for example when the slope is zero, according to this the R2 value is of about 0.4. The third and final approach that I have used is that described by Romdal et al. 2005. In which they use the second formula: R2=SSR/SSR+SSE, but they use (at least is how I understand it) the sum of squares of a regular OLS to estimate the sum of squares of the regression: so the corresponding code would be: #3 Red symbols in figure lm.y.x-lm(y~x) SSR-(deviance(lm(y~1))-sum((lm.y.x$residuals)^2)) SSE-sum((y-x)^2) R2-SSR/(SSR+SSE) This also of course stays within the expected range of 0 to 1, but has its own troubling behaivour, it does not peak at a slope of 1, there is an accelerated decrease at slopes less than 1, but not at slopes larger than 1, and it increases again at slopes less than 0 (like if negative associations between y and x would be better than a flat line, when the predictor is the same vairable as the dependen this does not make sense again). Any advice, recomendations for appropiate literature, or pieces of code would be highly appreciated. Best, Sebastian J. Sebastián Tello Department of Biological Sciences 285 Life Sciences Building Louisiana State University Baton Rouge, LA, 70803 (225) 578-4284 (office and lab.) __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] replace() error: new columns would leave holes after existing columns
Hi Kirsten, The second argument of replace() is actually an index, so it's trying to replace element number 2004-5-14 = 1985 with 134! See ?replace. I would need more of your code to understand your error exactly. I think you would be better off converting your dates to date format (?Date) and then using the julian() function to convert them to Julian. -- Kellie Wills Engineering Service Manager REvolution Computing On Fri, Oct 31, 2008 at 12:17 PM, Kirsten M Barrett [EMAIL PROTECTED]wrote: Hello, I have a problem with using replace() to convert a vector of dates from -mm-dd to julian date. For example, I type replace(x,2004-05-14,134) and I receive an error: Error in `[-.data.frame`(`*tmp*`, list, value = 134) : new columns would leave holes after existing columns If I can successfully convert, I have a script that will convert all of the dates in the vector (i.e., the same line repeated with the next -mm-dd and corresponding julian day). Any help would be much appreciated. Thank you, Kirsten Kirsten Barrett Mendenhall Postdoctoral Fellow, USGS Alaska Science Center 4210 University Drive Anchorage, AK [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] trellis: conditional statements within a panel function
hello, what i'm trying to achieve is shown in the comment (# ...) below, which is surely very wrong, but probably intuitively more meaningful that i can provide in words. xyplot(y~x2|x1+which,data=make.groups( a=data.a, b=data.b, c=data.c ), panel=function(...){ panel.abline(a=0,b=0) panel.points(...) # if(x1==0 which==a) panel.abline(...) }, layout=c(2,3), pch=20, ylim=c(-10,10), xlab=x2, cex=0.5 ) thank! [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Error Running TinnR with R
In the 'etc' directory of your installation path, e.g. C:\Program Files\R\etc Regards, Yihui -- Yihui Xie [EMAIL PROTECTED] Phone: +86-(0)10-82509086 Fax: +86-(0)10-82509086 Mobile: +86-15810805877 Homepage: http://www.yihui.name School of Statistics, Room 1037, Mingde Main Building, Renmin University of China, Beijing, 100872, China On Sat, Nov 1, 2008 at 9:36 AM, [EMAIL PROTECTED] wrote: I am sorry to be so slow but where do I find Rprofile.site? Thank you. Kevin Yihui Xie [EMAIL PROTECTED] wrote: Currently there are some problems with the building of Hmisc binary package; see CRAN for more information. My suggestion is you can remove (or comment out) the code lines not to use Hmisc for Tinn-R until it's available again on CRAN. As for me, I merely use two expressions in Rprofile.site and I have removed all the other code, as I don't need other functions of Tinn-R: options(pager=internal) .trPaths = c( 'C:/Documents and Settings/username/Application Data/Tinn-R/tmp/', 'C:/Documents and Settings/username/Application Data/Tinn-R/tmp/search.txt', 'C:/Documents and Settings/username/Application Data/Tinn-R/tmp/objects.txt', 'C:/Documents and Settings/username/Application Data/Tinn-R/tmp/file.r', 'C:/Documents and Settings/username/Application Data/Tinn-R/tmp/selection.r', 'C:/Documents and Settings/username/Application Data/Tinn-R/tmp/block.r', 'C:/Documents and Settings/username/Application Data/Tinn-R/tmp/lines.r') Regards, Yihui -- Yihui Xie [EMAIL PROTECTED] Phone: +86-(0)10-82509086 Fax: +86-(0)10-82509086 Mobile: +86-15810805877 Homepage: http://www.yihui.name School of Statistics, Room 1037, Mingde Main Building, Renmin University of China, Beijing, 100872, China On Fri, Oct 24, 2008 at 7:08 PM, bg708 [EMAIL PROTECTED] wrote: Hi, I managed to get it to work this way (changing the Rprofile.site file) with R2.7.2. I then made the mistake of updating to R2.8, and it stopped working. Now changing the Rprofile.site file does nothing, and sending a block of code gives me the Error in source(.trPaths[5], echo = T) : object .trPaths not found error message. I triple checked all the additional lines in the Rprofile.site file, path for .trPaths is correct. One point of note is that Hmisc package appears not to be available. Thanks for any help you could give me! Otherwise it might be wise to revert back to R2.7.2 or use the previous Tinn-R version. Bernardo __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] trellis: conditional statements within a panel function
On 10/31/08, erwann rogard [EMAIL PROTECTED] wrote: hello, what i'm trying to achieve is shown in the comment (# ...) below, which is surely very wrong, but probably intuitively more meaningful that i can provide in words. xyplot(y~x2|x1+which,data=make.groups( a=data.a, b=data.b, c=data.c ), panel=function(...){ panel.abline(a=0,b=0) panel.points(...) # if(x1==0 which==a) panel.abline(...) }, layout=c(2,3), pch=20, ylim=c(-10,10), xlab=x2, cex=0.5 ) Check ?which.packet. -Deepayan __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Help about the plot funtion
Try: ?locator On Fri, Oct 31, 2008 at 10:06 PM, German Urrus [EMAIL PROTECTED] wrote:  Hello, my name is German, I am from Argentina, and we are working at university with R. This work is about sounds and how they affect out ears, we are using R to analice different recordings of sounds and we need to get the intensity of this recordings in dB.  In order to get the intensity of the sounds, we are using the plot function, but we can not figure out what is the menaning of the graphic that we are getting, actually the values it takes are between -1 and 1. It would be very helpfull for us to now what this values are, and where we can get more information about the plot function, besides all the information that is in the help documentation. (we already have read it, and we could not find any help) Thanks for helping us Best regards Germán. Yahoo! Cocina Recetas prácticas y comida saludable http://ar.mujer.yahoo.com/cocina/ [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] count data with some conditions
Hi there, I have a data set: a=cbind(5,2,4,7,8,3,4,11,1,20) I want to count # of data, satistfying a[1]a[2:10]. Anyone helps me solving this case? Thank you in advance, Jin -- View this message in context: http://www.nabble.com/count-data-with-some-conditions-tp20275722p20275722.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.