[R] help with stepAIC and boot
Hello everybody. I've a problem with stepAIC inside a boot function (i'm trying to do bootstrap backward elimination, with a statistic function that allow one to specify the formula of a coxph model) For the moment the function would be aimed storing selected vars However I've a problem in the stepAIC code (coxph seems to me to be estimated)...the error message is Error in terms.formula(formula, special, data = data) : object 'db' not found aka stepAIC doesn't find bootstrap sample data How can i fix it? Many thanks in advance - library(boot) library(MASS) library(survival) hn - structure(list(pfs.time = c(309, 519, 318, 1114, 671, 1539, 1110, 1798, 2213, 1802, 1170, 1259, 189, 1939, 361, 1311, 1537, 706, 1934, 595, 173, 1147, 389, 854, 744, 172, 1530, 1993, 341, 1468, 540, 562, 1079, 1747, 1920, 1018, 899, 591, 1722, 400, 1099, 755, 1679, 1049, 1261, 408, 1695, 788, 324, 505, 505, 1159, 691, 924, 443, 781, 504, 1031, 205, 115, 506, 642, 637, 1426, 404, 944, 651, 863, 1758, 859, 502, 1899, 1746, 1517, 1087, 160, 1803, 706, 122, 996, 869, 611, 2484, 2438, 1095, 114, 574, 830, 2236, 616, 180, 398, 879, 281, 761, 764, 594, 874, 886, 1959, 1424, 239, 1650, 1689, 310, 1865, 2159, 1459, 980, 379, 1668, 394, 787, 1326, 1931, 2435, 356, 708, 1019, 100, 289), pfs.dummy = c(1, 0, 1, 0, 1, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 1, 1, 1, 0, 0, 1, 0, 0, 1, 1, 0, 0, 0, 1, 1, 0, 1, 0, 1, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 1, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 1, 0, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 1, 0, 1, 1), varA = c(12.8, 19, 15.4, 21.3, 18.6, 11.4, 0, 14.6, 7.9, 5.5, 23.5, 17.1, 13.2, 13.2, 30, 11, 12.8, 30.5, 13.6, 10.1, 13.5, 23.6, 11.1, 20.8, 6.1, 7, 13.3, 11.3, 13.9, 12, 8.3, 11.9, 5.2, 6.7, 28.4, 5.8, 22.5, 12.1, 12.4, 8.8, 20.2, 34.2, 5.1, 16, 24.7, 16.9, 0, 8.1, 0, 17.6, 6.5, 16.5, 22.6, 18.3, 12.1, 16.2, 14.9, 21.3, 15.9, 2.9, 9.9, 13.3, 0, 12.8, 19.9, 17.6, 21.8, 15.9, 20.5, 13.4, 14.1, 13, 5.4, 16.3, 18.7, 13, 21, 3.7, 23.6, 17.1, 6.1, 7.4, 0, 14.9, 11.5, 29.8, 8.8, 14, 13.1, 5.5, 16.4, 11.4, 6.3, 18.1, 11.9, 14.2, 25.8, 9.6, 5.3, 22.1, 12.5, 25.6, 17.5, 7.2, 5.7, 12.8, 11.6, 22.7, 17.4, 23.7, 6.7, 10.3, 13.6, 5, 16.4, 16.1, 20.5, 15.6, 6.7, 20, 14.7), varB = c(3.3, 4.13, 14.25, 5.2, 13.25, 5, 0, 19.1, 7.2, 24.2, 12.6, 11.1, 21.32, 13.6, 36.9, 9.98, 8.51, 9.4, 8.12, 3, 31.25, 10.3, 13.1, 9.4, 6.94, 10.9, 8.6, 6.2, 27, 2.5, 12.6, 17.2, 13.2, 4.6, 17.6, 3.3, 6.7, 4.9, 12.5, 5.4, 3.9, 11.7, 0.6, 26.3, 23.9, 15.9, 0, 13.1, 0, 4.45, 5.2, 18.8, 7.73, 12.4, 4.2, 2.71, 3.4, 8.5, 15.6, 18.4, 6.72, 31.15, 0, 8.1, 25.7, 5.8, 48.7, 7.43, 6.1, 9.4, 3.9, 15.1, 5.2, 7.2, 6.8, 7.4, 2.5, 6.16, 32.9, 6.5, 15.5, 3.96, 0, 4.2, 4.94, 14.7, 7.4, 8.6, 7.2, 20.5, 5.6, 8.75, 5.6, 10.2, 2.7, 7.4, 9.9, 6.14, 2.2, 36, 2.4, 4.6, 5.9, 4.9, 21.9, 9.1, 2.5, 33.2, 11.4, 5, 2.2, 2.3, 14.9, 6, 13.9, 4.7, 30.44, 3.4, 10.7, 7, 7.9), varC = c(24.6, 48.5, 135.1, 67.9, 156, 34.3, 0, 14.6, 33.2, 75.6, 192.8, 118.4, 174.1, 113.5, 699.4, 63.1, 68, 175.1, 66.3, 17.1, 265.6, 162.1, 87.1, 122.5, 23.3, 40.8, 65.1, 46, 241.6, 17.1, 64.5, 123.9, 37.9, 17.7, 284.1, 10.3, 95.3, 42.6, 92.2, 26.8, 45.2, 238.3, 2.3, 261, 380.4, 147.4, 0, 57, 0, 45.6, 19.1, 186.8, 107.9, 143.3, 27.9, 27.8, 30.1, 110.1, 161.3, 32.2, 42, 249, 0, 54.7, 313.1, 63.8, 656, 80, 81.5, 78.4, 31.4, 110.3, 14.9, 76.1, 78.5, 59.8, 31.7, 12.1, 457.9, 62.5, 49.5, 16, 0, 37, 345, 255, 35.9, 72, 54.5, 61.8, 55.9, 57.2, 18.4, 104.2, 18.2, 66.1, 156.1, 35.4, 7.2, 523.8, 17.7, 68.4, 52.8, 19.7, 74.6, 72.5, 19.2, 498.7, 121.7, 78.4, 10.8, 14, 133.2, 17.3, 133.9, 43.8, 375.2, 32.9, 44.3, 86.5, 63.8), varD = c(4.3, 10.3, 4.7, 5.4, 6.7, 2.7, 7, 9.2, 17.7, 0, 4.6, 0, 12.1, 5.5, 0, 3.4, 7, 7.2, 10.8, 8.3, 13.7, 19.6, 0, 11.8, 0, 7.5, 0, 0, 3.2, 0, 6.2, 9.6, 0, 4, 5.1, 0, 18.8, 16.1, 6.4, 0, 12.2, 15.8, 10.1, 13.9, 0, 11.7, 15.5, 0, 13.1, 11, 0, 0, 32.7, 6.3, 5.9, 0, 0, 12.8, 0, 0, 3.8, 7.3, 3.7, 18.1, 15.6, 14.9, 21.4, 13, 0, 8.1, 9.6, 8.1, 0, 13.1, 12.1, 10.7, 8.5, 2.9, 0, 15.4, 0, 3.3, 4.8, 0, 7.5, 38.9, 3.6, 6.7, 5.7, 4.4, 18.9, 11.5, 0, 15.2, 0, 8.5, 0, 6.4, 0, 17.6, 0, 26.7, 4.6, 4.4, 0, 5.9, 9.2, 20.2, 0, 0, 10.1, 7, 11.3, 0, 4.4, 10.2, 11.2, 0, 0, 0, 4), varE = c(1.9, 3.55, 1.96, 3.1, 1.5, 0, 8.3, 0.7, 4.3, 0, 2.1, 0, 11.69, 2, 0, 2.5, 4.4, 0.8, 0.8, 9.4, 19.19, 14.2, 0, 2.3, 0, 6.7, 0, 0, 1.7, 0, 1.6, 8.3, 0, 20.6, 1.8, 0, 12.5, 15.1, 1.6, 0, 1.6, 2, 25.3, 4.2, 0, 6.55, 16.7, 0, 8.7, 4.84, 0, 0, 16.72, 8.4, 1.5, 0, 0, 3.8, 0, 0, 3.89, 1.5, 4, 15.65, 4.3, 2.5, 10.9, 3.23, 0, 2.9, 7.7, 8.3, 0, 15.6, 5.2, 3.9, 1.6, 3.52, 0, 5.4, 0, 4.3, 2.7, 0, 8, 5.1, 0.9, 2.35, 0.6, 14.2, 10.7, 8.7, 0, 10.9, 0, 2.3, 0, 1.8, 0, 8.7, 0, 16.2, 0.1, 2.8, 0, 0.2, 8.3, 7.3, 0, 0, 10.3, 0.6, 7.1, 0, 5.3, 8.2, 0.83, 0, 0, 0, 1.8 ), varF = c(4.5, 21.4, 5.1, 9.4, 6.1, 0, 33.6, 5.2, 47.4, 0, 7.1, 0,
Re: [R] Building R for better performance
Jonathan, I myself tried something like this - comparing gcc, clang and intel on a Mac. From my experiences in HPC on the university cluster (where we also use the Xeon Phi, Landeshochleistungscluster University RWTH Aachen), the Intel compiler has better code optimization in regard to vectorisation, etc. (clang is up to now suffering from a not yet implemented OpenMP library). Here is a revolutionanalytics article about this topic: http://blog.revolutionanalytics.com/2010/06/performance-benefits-of-multithreaded-r.html As I usually use the Rcpp package for C++ extensions this could give me further performance. Though, I already failed when trying to compile R with the Intel compiler and linking against the MKL (see my topic in the Intel developer zone: http://software.intel.com/en-us/comment/1767418 and my threads on the R-User list: https://stat.ethz.ch/pipermail/r-sig-mac/2013-November/010472.html). So, to your questions: 1) I think that most admins do not even use the Intel compiler to compile R - this seems to me rare. There are some people I know they do and I think they could be aware of it - but these are only a few. As R is growing in usage and I do know from regional user meetings that very large companies start using it in their BI units - this should be of interest. 2) I would really welcome this step because compilation with intel (especially on a Mac) and linking to the MKL seems to be delicate. I am interested in the data - so if it is possible send it via the list or directly to my account. Further, could you show some code that you used for the computations? Best Simon On 04 Mar 2014, at 22:44, Anspach, Jonathan P jonathan.p.ansp...@intel.com wrote: Greetings, I'm a software engineer with Intel. Recently I've been investigating R performance on Intel Xeon and Xeon Phi processors and RH Linux. I've also compared the performance of R built with the Intel compilers and Intel Math Kernel Library to a default build (no config options) that uses the GNU compilers. To my dismay, I've found that the GNU build always runs on a single CPU core, even during matrix operations. The Intel build runs matrix operations on multiple cores, so it is much faster on those operations. Running the benchmark-2.5 on a 24 core Xeon system, the Intel build is 13x faster than the GNU build (21 seconds vs 275 seconds). Unfortunately, this advantage is not documented anywhere that I can see. Building with the Intel tools is very easy. Assuming the tools are installed in /opt/intel/composerxe, the process is simply (in bash shell): $ . /opt/intel/composerxe/bin/compilervars.sh intel64 $ ./configure --with-blas=-L/opt/intel/composerxe/mkl/lib/intel64 -lmkl_intel_lp64 -lmkl_intel_thread -lmkl_core -liomp5 -lpthread -lm --with-lapack CC=icc CFLAGS=-O2 CXX=icpc CXXFLAGS=-O2 F77=ifort FFLAGS=-O2 FC=ifort FCFLAGS=-O2 $ make $ make check My questions are: 1) Do most system admins and/or R installers know about this performance difference, and use the Intel tools to build R? 2) Can we add information on the advantage of building with the Intel tools, and how to do it, to the installation instructions and FAQ? I can post my data if anyone is interested. Thanks, Jonathan Anspach Sr. Software Engineer Intel Corp. jonathan.p.ansp...@intel.com 713-751-9460 __ 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] Having a plot with points and line with different colors
Hi experts I want to have a plot, which consist of line and points something like this: plot(data, type=o) but I would like that lines and point have different colors. for example line should be blue and points should be black how can I do that in R? [[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] Having a plot with points and line with different colors
On 03/05/2014 08:59 PM, Baro wrote: Hi experts I want to have a plot, which consist of line and points something like this: plot(data, type=o) but I would like that lines and point have different colors. for example line should be blue and points should be black how can I do that in R? Hi Baro, There are a few ways to do this, here are two. Probably the easiest is to plot the lines, then add the symbols with points, allowing the line to have one color and the point another. You can overplot the o type of plot with points of a different color, but this occasionally leads to color bleed on the points if the overplotting is not exact. Jim __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Having a plot with points and line with different colors
thanks Jim, could you please show me some example? On Wed, Mar 5, 2014 at 2:02 AM, Jim Lemon j...@bitwrit.com.au wrote: On 03/05/2014 08:59 PM, Baro wrote: Hi experts I want to have a plot, which consist of line and points something like this: plot(data, type=o) but I would like that lines and point have different colors. for example line should be blue and points should be black how can I do that in R? Hi Baro, There are a few ways to do this, here are two. Probably the easiest is to plot the lines, then add the symbols with points, allowing the line to have one color and the point another. You can overplot the o type of plot with points of a different color, but this occasionally leads to color bleed on the points if the overplotting is not exact. Jim [[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] Having a plot with points and line with different colors
On 03/05/2014 09:09 PM, Baro wrote: thanks Jim, could you please show me some example? # plot lines, then points plot(1:5,type=l,col=blue) points(1:5,col=black) # plot both, then overplot points plot(1:5,type=o,col=blue) points(1:5,col=black) Jim __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Is this a mistake in 'An Introduction to R'?
On 04 Mar 2014, at 21:21 , Geoff Loveman ge...@lovemans.co.uk wrote: In 'An Introduction to R', section 11.7 on nonlinear least squares fitting, the following example is given for obtaining the standard errors of the estimated parameters: To obtain the approximate standard errors (SE) of the estimates we do: sqrt(diag(2*out$minimum/(length(y) - 2) * solve(out$hessian)))The 2 in the line above represents the number of parameters. I know the inverted Hessian is multiplied by the mean square error and that the denominator of the MSE is the degrees of freedom (number of samples - number of parameters) but why does the numerator of the MSE (which is the RSS) get multiplied by the number of parameters? I have read through explanations of the method for obtaining the SE but I don't see where the MSE gets multiplied by the number of parameters or why this is needed as shown in the example? There are two 2's in that line, and I'd expect that only the last one has to do with the number of parameters, and the other one has to do with whether the Hessian is the second derivative of the sum of squares or of the negative loglikelihood function (half the sum of squares). Quick check: In a linear model, we have ssd = || Y- X beta ||^2 gradient = -2 (Y - X beta )'X Hessian H = 2 X'X and as we know, V(beta) = sigma^2 (X'X)^-1 = 2 sigma^2 H^-1 -pd Thanks for any help! Geoff Loveman Tech lead SMERAS QQ Maritime Life Support -- View this message in context: http://r.789695.n4.nabble.com/Is-this-a-mistake-in-An-Introduction-to-R-tp4686217.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. -- Peter Dalgaard, Professor Center for Statistics, Copenhagen Business School Solbjerg Plads 3, 2000 Frederiksberg, Denmark Phone: (+45)38153501 Email: pd@cbs.dk Priv: pda...@gmail.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] Sweave: cat() in a chunk with option results=tex doesn't producelinebreaks at the end of a character string anymore
Hello, everyone, I am struggling with an Sweave-problem that didn't occur sofar (and I have no clue what I might have changed in my system; see below). The following example *.Rnw file's only task is (for simplicity) to output text with a little bit of TeX-code with linebreaks (e. g., to be better readable): \documentclass{article} \begin{document} Text 1 (not through cat()) followed by a double-backslash and a blank line, i.e., a linebreak.\\ echo=FALSE, results=tex= cat( Text 2 (through cat()) with a respected newline here\n, and likewise two tabs here\t\t followed by two double-backslashes , and IGNORED newline escape characters at the end. \n \n, sep = ) cat( Text 3 (through cat()) just to finish this example and , with also IGNORED newline escapes at its end. \n \n, sep = ) @ \end{document} When processed by Sweave I obtain the following TeX file: \documentclass{article} \usepackage{Sweave} \begin{document} Text 1 (not through cat()) followed by a double-backslash and a blank line, i.e., a linebreak.\\ Text 2 (through cat()) with a respected newline here and likewise two tabs here followed by two double-backslashes and IGNORED newline escape characters at the end.\\ Text 3 (through cat()) to finish this example and with also IGNORED newline escapes at its end. \end{document} the point being that the linebreaks (\n) at the ends of the output character strings have been completely ignored. I wasn't successful in searching the archive, RSeek, SE and the like. Any idea where and how to continue to search, or what to do to get back Sweave's old behavior (except for reinstalling R ... which of course is still an option)? Thanks for any hint! Best regards -- Gerrit PS: sessionInfo() R version 3.0.2 (2013-09-25) Platform: x86_64-w64-mingw32/x64 (64-bit) locale: [1] LC_COLLATE=German_Germany.1252 LC_CTYPE=German_Germany.1252 [3] LC_MONETARY=German_Germany.1252 LC_NUMERIC=C [5] LC_TIME=German_Germany.1252 attached base packages: [1] graphics grDevices utils datasets stats methods base other attached packages: [1] fortunes_1.5-2 loaded via a namespace (and not attached): [1] grid_3.0.2 lattice_0.20-27 tools_3.0.2 - Dr. Gerrit Eichner Mathematical Institute, Room 212 gerrit.eich...@math.uni-giessen.de Justus-Liebig-University Giessen Tel: +49-(0)641-99-32104 Arndtstr. 2, 35392 Giessen, Germany Fax: +49-(0)641-99-32109http://www.uni-giessen.de/cms/eichner __ 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] Sweave: cat() in a chunk with option results=tex doesn't producelinebreaks at the end of a character string anymore
On 05/03/2014 7:32 AM, Gerrit Eichner wrote: Hello, everyone, I am struggling with an Sweave-problem that didn't occur sofar (and I have no clue what I might have changed in my system; see below). The following example *.Rnw file's only task is (for simplicity) to output text with a little bit of TeX-code with linebreaks (e. g., to be better readable): \documentclass{article} \begin{document} Text 1 (not through cat()) followed by a double-backslash and a blank line, i.e., a linebreak.\\ echo=FALSE, results=tex= cat( Text 2 (through cat()) with a respected newline here\n, and likewise two tabs here\t\t followed by two double-backslashes , and IGNORED newline escape characters at the end. \n \n, sep = ) cat( Text 3 (through cat()) just to finish this example and , with also IGNORED newline escapes at its end. \n \n, sep = ) @ \end{document} When processed by Sweave I obtain the following TeX file: \documentclass{article} \usepackage{Sweave} \begin{document} Text 1 (not through cat()) followed by a double-backslash and a blank line, i.e., a linebreak.\\ Text 2 (through cat()) with a respected newline here and likewise two tabs here followed by two double-backslashes and IGNORED newline escape characters at the end.\\ Text 3 (through cat()) to finish this example and with also IGNORED newline escapes at its end. \end{document} the point being that the linebreaks (\n) at the ends of the output character strings have been completely ignored. See the strip.white option in ?RweaveLatex. Duncan Murdoch I wasn't successful in searching the archive, RSeek, SE and the like. Any idea where and how to continue to search, or what to do to get back Sweave's old behavior (except for reinstalling R ... which of course is still an option)? Thanks for any hint! Best regards -- Gerrit PS: sessionInfo() R version 3.0.2 (2013-09-25) Platform: x86_64-w64-mingw32/x64 (64-bit) locale: [1] LC_COLLATE=German_Germany.1252 LC_CTYPE=German_Germany.1252 [3] LC_MONETARY=German_Germany.1252 LC_NUMERIC=C [5] LC_TIME=German_Germany.1252 attached base packages: [1] graphics grDevices utils datasets stats methods base other attached packages: [1] fortunes_1.5-2 loaded via a namespace (and not attached): [1] grid_3.0.2 lattice_0.20-27 tools_3.0.2 - Dr. Gerrit Eichner Mathematical Institute, Room 212 gerrit.eich...@math.uni-giessen.de Justus-Liebig-University Giessen Tel: +49-(0)641-99-32104 Arndtstr. 2, 35392 Giessen, Germany Fax: +49-(0)641-99-32109http://www.uni-giessen.de/cms/eichner __ 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] Sampling according to type
I have a matrix where each entry represents a data subject's type, 1 or 0: n - 10 ntype - rbinom(n, 1, 0.5) and I'd like to sample say 3 subjects from ntype where those subjects who are Type 1 are selected with probability say 0.75, and Type 0 with (1-0.75). (So the sample would produce a list with three indices each referring to a position within ntype.) Can anyone suggest a way to do this please? Thank you, Thomas Chesney This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it. Please do not use, copy or disclose the information contained in this message or in any attachment. Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham. This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation. __ 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] Sweave: cat() in a chunk with option results=texdoesn'tproducelinebreaks at the end of a character string anymore
Thanks, Duncan, but, sorry, including \SweaveOpts{strip.white=true} doesn't help. Have tried that before and forgot to mention; sorry! BTW: using \SweaveOpts{strip.white=true}, i.e., with quotation marks as requested in RweaveLatex()'s help file, throws out Error in match.arg(options$strip.white, c(true, false, all)) : 'arg' should be one of true, false, all which used to happen before (on my system) so that I used to use it w/o quotes successfully. (!?) Still big question marks ... Any further ideas? Regards -- Gerrit On Wed, 5 Mar 2014, Duncan Murdoch wrote: On 05/03/2014 7:32 AM, Gerrit Eichner wrote: Hello, everyone, I am struggling with an Sweave-problem that didn't occur sofar (and I have no clue what I might have changed in my system; see below). The following example *.Rnw file's only task is (for simplicity) to output text with a little bit of TeX-code with linebreaks (e. g., to be better readable): \documentclass{article} \begin{document} Text 1 (not through cat()) followed by a double-backslash and a blank line, i.e., a linebreak.\\ echo=FALSE, results=tex= cat( Text 2 (through cat()) with a respected newline here\n, and likewise two tabs here\t\t followed by two double-backslashes , and IGNORED newline escape characters at the end. \n \n, sep = ) cat( Text 3 (through cat()) just to finish this example and , with also IGNORED newline escapes at its end. \n \n, sep = ) @ \end{document} When processed by Sweave I obtain the following TeX file: \documentclass{article} \usepackage{Sweave} \begin{document} Text 1 (not through cat()) followed by a double-backslash and a blank line, i.e., a linebreak.\\ Text 2 (through cat()) with a respected newline here and likewise two tabs here followed by two double-backslashes and IGNORED newline escape characters at the end.\\ Text 3 (through cat()) to finish this example and with also IGNORED newline escapes at its end. \end{document} the point being that the linebreaks (\n) at the ends of the output character strings have been completely ignored. See the strip.white option in ?RweaveLatex. Duncan Murdoch I wasn't successful in searching the archive, RSeek, SE and the like. Any idea where and how to continue to search, or what to do to get back Sweave's old behavior (except for reinstalling R ... which of course is still an option)? Thanks for any hint! Best regards -- Gerrit PS: sessionInfo() R version 3.0.2 (2013-09-25) Platform: x86_64-w64-mingw32/x64 (64-bit) locale: [1] LC_COLLATE=German_Germany.1252 LC_CTYPE=German_Germany.1252 [3] LC_MONETARY=German_Germany.1252 LC_NUMERIC=C [5] LC_TIME=German_Germany.1252 attached base packages: [1] graphics grDevices utils datasets stats methods base other attached packages: [1] fortunes_1.5-2 loaded via a namespace (and not attached): [1] grid_3.0.2 lattice_0.20-27 tools_3.0.2 - Dr. Gerrit Eichner Mathematical Institute, Room 212 gerrit.eich...@math.uni-giessen.de Justus-Liebig-University Giessen Tel: +49-(0)641-99-32104 Arndtstr. 2, 35392 Giessen, Germany Fax: +49-(0)641-99-32109http://www.uni-giessen.de/cms/eichner __ 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] Sweave: cat() in a chunk with option results=texdoesn'tproducelinebreaks at the end of a character string anymore
On 05/03/2014 9:29 AM, Gerrit Eichner wrote: Thanks, Duncan, but, sorry, including \SweaveOpts{strip.white=true} doesn't help. Have tried that before and forgot to mention; sorry! true is the default. You want false. Duncan Murdoch BTW: using \SweaveOpts{strip.white=true}, i.e., with quotation marks as requested in RweaveLatex()'s help file, throws out Error in match.arg(options$strip.white, c(true, false, all)) : 'arg' should be one of true, false, all which used to happen before (on my system) so that I used to use it w/o quotes successfully. (!?) Still big question marks ... Any further ideas? Regards -- Gerrit On Wed, 5 Mar 2014, Duncan Murdoch wrote: On 05/03/2014 7:32 AM, Gerrit Eichner wrote: Hello, everyone, I am struggling with an Sweave-problem that didn't occur sofar (and I have no clue what I might have changed in my system; see below). The following example *.Rnw file's only task is (for simplicity) to output text with a little bit of TeX-code with linebreaks (e. g., to be better readable): \documentclass{article} \begin{document} Text 1 (not through cat()) followed by a double-backslash and a blank line, i.e., a linebreak.\\ echo=FALSE, results=tex= cat( Text 2 (through cat()) with a respected newline here\n, and likewise two tabs here\t\t followed by two double-backslashes , and IGNORED newline escape characters at the end. \n \n, sep = ) cat( Text 3 (through cat()) just to finish this example and , with also IGNORED newline escapes at its end. \n \n, sep = ) @ \end{document} When processed by Sweave I obtain the following TeX file: \documentclass{article} \usepackage{Sweave} \begin{document} Text 1 (not through cat()) followed by a double-backslash and a blank line, i.e., a linebreak.\\ Text 2 (through cat()) with a respected newline here and likewise two tabs here followed by two double-backslashes and IGNORED newline escape characters at the end.\\ Text 3 (through cat()) to finish this example and with also IGNORED newline escapes at its end. \end{document} the point being that the linebreaks (\n) at the ends of the output character strings have been completely ignored. See the strip.white option in ?RweaveLatex. Duncan Murdoch I wasn't successful in searching the archive, RSeek, SE and the like. Any idea where and how to continue to search, or what to do to get back Sweave's old behavior (except for reinstalling R ... which of course is still an option)? Thanks for any hint! Best regards -- Gerrit PS: sessionInfo() R version 3.0.2 (2013-09-25) Platform: x86_64-w64-mingw32/x64 (64-bit) locale: [1] LC_COLLATE=German_Germany.1252 LC_CTYPE=German_Germany.1252 [3] LC_MONETARY=German_Germany.1252 LC_NUMERIC=C [5] LC_TIME=German_Germany.1252 attached base packages: [1] graphics grDevices utils datasets stats methods base other attached packages: [1] fortunes_1.5-2 loaded via a namespace (and not attached): [1] grid_3.0.2 lattice_0.20-27 tools_3.0.2 - Dr. Gerrit Eichner Mathematical Institute, Room 212 gerrit.eich...@math.uni-giessen.de Justus-Liebig-University Giessen Tel: +49-(0)641-99-32104 Arndtstr. 2, 35392 Giessen, Germany Fax: +49-(0)641-99-32109http://www.uni-giessen.de/cms/eichner __ 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] Sampling according to type
If I understood correctly, you need weighted sampling. Try 'prob' argument from 'sample'. For your example: n - 10 ntype - rbinom(n, 1, 0.5) myProbs - rep(1/10, 10) # equally likely myProbs[ which(ntype == 0)] - 0.75/7 # Divide so the sum will be 1.0 myProbs[ which(ntype == 1)] - 0.25/3 sample(ntype,3, prob=myProbs) On 5 March 2014 15:20, Thomas thomas.ches...@nottingham.ac.uk wrote: I have a matrix where each entry represents a data subject's type, 1 or 0: n - 10 ntype - rbinom(n, 1, 0.5) and I'd like to sample say 3 subjects from ntype where those subjects who are Type 1 are selected with probability say 0.75, and Type 0 with (1-0.75). (So the sample would produce a list with three indices each referring to a position within ntype.) Can anyone suggest a way to do this please? Thank you, Thomas Chesney This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it. Please do not use, copy or disclose the information contained in this message or in any attachment. Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham. This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation. __ 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] Sampling according to type
myProbs[ which(ntype == 0)] - 0.75/7 # Divide so the sum will be 1.0 myProbs[ which(ntype == 1)] - 0.25/3 Here of course you need to divide by number of 0s and 1s, 7 and 3 were was just an example. __ 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] Having a plot with points and line with different colors
Alternatively use type=c instead of l which will create line breaks where the points will go. plot(1:5,type=c, col=blue) points(1:5,col=black) David -Original Message- From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On Behalf Of Jim Lemon Sent: Wednesday, March 5, 2014 4:08 AM To: Baro Cc: R help Subject: Re: [R] Having a plot with points and line with different colors On 03/05/2014 09:09 PM, Baro wrote: thanks Jim, could you please show me some example? # plot lines, then points plot(1:5,type=l,col=blue) points(1:5,col=black) # plot both, then overplot points plot(1:5,type=o,col=blue) points(1:5,col=black) Jim __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-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] Sweave: cat() in a chunk with option results=texdoesn'tproducelinebreaksatthe end of a character string anymore
Thanks, Duncan, I got it! I missread/-understood strip.white's meaning and thought it would refer only to blank lines within real code chunks (i.e., to those between \begin{Schunk} and \end{Schunk}), but not to the (TeX-)output which was generated by code chunks with results=tex. My goal: I don't want blank lines at the beginning or end of real R output between \begin{Schunk} and \end{Schunk}), but want to be able to create linebreaks using cat( \n) explicitly in chunks with results=tex The solution appears to be: use strip.white=false locally in case of need, i.e., as the option for the respective code chunks (which also have results=tex, and from which I want only linebreaks in the TeX-output). Actually, embarrassingly simple. Thanks again -- Gerrit On Wed, 5 Mar 2014, Duncan Murdoch wrote: On 05/03/2014 9:29 AM, Gerrit Eichner wrote: Thanks, Duncan, but, sorry, including \SweaveOpts{strip.white=true} doesn't help. Have tried that before and forgot to mention; sorry! true is the default. You want false. Duncan Murdoch BTW: using \SweaveOpts{strip.white=true}, i.e., with quotation marks as requested in RweaveLatex()'s help file, throws out Error in match.arg(options$strip.white, c(true, false, all)) : 'arg' should be one of true, false, all which used to happen before (on my system) so that I used to use it w/o quotes successfully. (!?) Still big question marks ... Any further ideas? Regards -- Gerrit On Wed, 5 Mar 2014, Duncan Murdoch wrote: On 05/03/2014 7:32 AM, Gerrit Eichner wrote: Hello, everyone, I am struggling with an Sweave-problem that didn't occur sofar (and I have no clue what I might have changed in my system; see below). The following example *.Rnw file's only task is (for simplicity) to output text with a little bit of TeX-code with linebreaks (e. g., to be better readable): \documentclass{article} \begin{document} Text 1 (not through cat()) followed by a double-backslash and a blank line, i.e., a linebreak.\\ echo=FALSE, results=tex= cat( Text 2 (through cat()) with a respected newline here\n, and likewise two tabs here\t\t followed by two double-backslashes , and IGNORED newline escape characters at the end. \n \n, sep = ) cat( Text 3 (through cat()) just to finish this example and , with also IGNORED newline escapes at its end. \n \n, sep = ) @ \end{document} When processed by Sweave I obtain the following TeX file: \documentclass{article} \usepackage{Sweave} \begin{document} Text 1 (not through cat()) followed by a double-backslash and a blank line, i.e., a linebreak.\\ Text 2 (through cat()) with a respected newline here and likewise two tabs here followed by two double-backslashes and IGNORED newline escape characters at the end.\\ Text 3 (through cat()) to finish this example and with also IGNORED newline escapes at its end. \end{document} the point being that the linebreaks (\n) at the ends of the output character strings have been completely ignored. See the strip.white option in ?RweaveLatex. Duncan Murdoch I wasn't successful in searching the archive, RSeek, SE and the like. Any idea where and how to continue to search, or what to do to get back Sweave's old behavior (except for reinstalling R ... which of course is still an option)? Thanks for any hint! Best regards -- Gerrit PS: sessionInfo() R version 3.0.2 (2013-09-25) Platform: x86_64-w64-mingw32/x64 (64-bit) locale: [1] LC_COLLATE=German_Germany.1252 LC_CTYPE=German_Germany.1252 [3] LC_MONETARY=German_Germany.1252 LC_NUMERIC=C [5] LC_TIME=German_Germany.1252 attached base packages: [1] graphics grDevices utils datasets stats methods base other attached packages: [1] fortunes_1.5-2 loaded via a namespace (and not attached): [1] grid_3.0.2 lattice_0.20-27 tools_3.0.2 - Dr. Gerrit Eichner Mathematical Institute, Room 212 gerrit.eich...@math.uni-giessen.de Justus-Liebig-University Giessen Tel: +49-(0)641-99-32104 Arndtstr. 2, 35392 Giessen, Germany Fax: +49-(0)641-99-32109http://www.uni-giessen.de/cms/eichner __ 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] Having a plot with points and line with different colors
Thanks a lot for your helps On Wed, Mar 5, 2014 at 7:53 AM, David Carlson dcarl...@tamu.edu wrote: Alternatively use type=c instead of l which will create line breaks where the points will go. plot(1:5,type=c, col=blue) points(1:5,col=black) David -Original Message- From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On Behalf Of Jim Lemon Sent: Wednesday, March 5, 2014 4:08 AM To: Baro Cc: R help Subject: Re: [R] Having a plot with points and line with different colors On 03/05/2014 09:09 PM, Baro wrote: thanks Jim, could you please show me some example? # plot lines, then points plot(1:5,type=l,col=blue) points(1:5,col=black) # plot both, then overplot points plot(1:5,type=o,col=blue) points(1:5,col=black) Jim __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[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] histograms embedded in a plot (as alternative to jitter)
Dear R list, I am plotting a discrete valued number on the y axis against a continuous variable on the x axis. To allow sample size to be viewed for the discrete groups I am using vertical jitter. So my code is along the lines of y-rpois(500,2) x-rnorm(500,y,1) plot(x,jitter(y)) It has not been suggested that a more informative view of the sample size might be if a histogram could be inserted into the plot (instead of the jittered rows of points) for the values at each y-value Many thanks if anyone can think of a way to do this This email may contain information that is privileged, confidential or otherwise protected from disclosure. It must not be used by, or its contents copied or disclosed to, persons other than the addressee. If you have received this email in error please notify the sender immediately and delete the email. This message has been scanned for viruses.\ \ \ [[alter...{{dropped:6}} __ 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] margins and device: different results with pdf() and svg()
The following works fine with either the plot window or a pdf device: pdf(test.pdf,width=12,height=8) par(mar=c(5,4,4,5)) plot(421:450,1:30,xlim=c(300,1100), ylim=c(0,100), type=l,col=blue,xlab=Wavelength,ylab=Transmittance %) lines(401:1000, (401:1000)/10,col=green) par(new=TRUE) axis(4) mtext(Reflectance,side=4,line=3) title(Filters MCA01) dev.off() but if I use svg() I get a wrong plot. Why? Any way to get the same plot in svg as in pdf? Thanks Agus sessionInfo() R version 3.0.2 (2013-09-25) Platform: x86_64-pc-linux-gnu (64-bit) locale: [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C LC_TIME=es_ES.UTF-8LC_COLLATE=en_US.UTF-8 [5] LC_MONETARY=es_ES.UTF-8LC_MESSAGES=en_US.UTF-8 LC_PAPER=es_ES.UTF-8 LC_NAME=C [9] LC_ADDRESS=C LC_TELEPHONE=C LC_MEASUREMENT=es_ES.UTF-8 LC_IDENTIFICATION=C attached base packages: [1] stats graphics grDevices utils datasets methods base other attached packages: [1] MASS_7.3-29 reshape_0.8.4 plyr_1.8ggplot2_0.9.3.1 raster_2.2-12 rgdal_0.8-14sp_1.0-14 loaded via a namespace (and not attached): [1] colorspace_1.2-4 dichromat_2.0-0digest_0.6.4 grid_3.0.2 gtable_0.1.2 labeling_0.2 [7] lattice_0.20-24munsell_0.4.2 proto_0.3-10 RColorBrewer_1.0-5 reshape2_1.2.2 scales_0.2.3 [13] stringr_0.6.2 tools_3.0.2 -- Agustin Lobo aloboa...@gmail.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] pdf: plotting very small points
Dear all, Ghostscript (version 9.05) cannot open pdfs produced with R (version 3.0.2) that contain very small points in plots. Other pdf readers (evince, AcrobatReader) are able to open the file. If setting cex=0, the pdf is readable again: # does open with gs pdf(cex_small.pdf) plot(1,1,cex=10^-3) dev.off() # does not open with gs pdf(cex_verysmall.pdf) plot(1,1,cex=10^-4) dev.off() # does open with gs pdf(cex_zero.pdf) plot(1,1,cex=0) dev.off() For sure, I can check the value assigned to cex before plotting. However, this is rather inconvenient. I need to convert the graphs using convert to other file formats, hence simply using another pdf readers is not an option either. Any suggestions? Cheers, Rebecca Hiller [[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] To obtain a graph by geom_bar or other
Hi I have the dataframe following : data.frame(Annee =c(rep(2004,200), rep(2005,200)), CiradMed = c(rep(Cirad,100), rep(Med, 100), rep(Cirad,100), rep(Med, 100)) , Type=c(rep(T1, 25), rep(T2, 20), rep(T3, 30), rep(T4, 25), rep(T1, 20), rep(T2, 30), rep(T3, 40), rep(T4, 10), rep(T1, 25), rep(T2, 30), rep(T3, 35), rep(T4, 10), rep(T1, 15), rep(T2, 35), rep(T3, 40), rep(T4, 10)) , stringsAsFactors=FALSE) Is it possible to build with ggplot the same graph obtained by graphics packages (see the joined file) ? I tried by using geom_bar(position=dodge) but it is not correct. Thank you for your help Michel __ 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] Script question
I have a script that works fine if I copy the whole thing from from a text editor and paste it into R, but doesn't execute properly if the file is called by the source command. Any clues on how to fix this? I can post the script if necessary. o M. Omar Faison, PhD Director, Office of Sponsored Programs and Research Virginia State University 804-631-3330 http://twitter.com/omarfaison âThe information in this email and any attachments may be confidential and privileged. Access to this email by anyone other than the intended addressee is unauthorized. If you are not the intended recipient (or the employee or agent responsible for delivering this information to the intended recipient) please notify the sender by reply email and immediately delete this email and any copies from your computer and/or storage system. The sender does not authorize the use, distribution, disclosure or reproduction of this email (or any part of its contents) by anyone other than the intended recipient(s). No representation is made that this email and any attachments are free of viruses. Virus scanning is recommended and is the responsibility of the recipient.â [[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] pdf: plotting very small points
On 05/03/2014 8:34 AM, rebecca.hil...@meteoswiss.ch wrote: Dear all, Ghostscript (version 9.05) cannot open pdfs produced with R (version 3.0.2) that contain very small points in plots. Other pdf readers (evince, AcrobatReader) are able to open the file. If setting cex=0, the pdf is readable again: # does open with gs pdf(cex_small.pdf) plot(1,1,cex=10^-3) dev.off() # does not open with gs pdf(cex_verysmall.pdf) plot(1,1,cex=10^-4) dev.off() # does open with gs pdf(cex_zero.pdf) plot(1,1,cex=0) dev.off() For sure, I can check the value assigned to cex before plotting. However, this is rather inconvenient. I need to convert the graphs using convert to other file formats, hence simply using another pdf readers is not an option either. Any suggestions? From your description, it sounds like a Ghostscript bug, so you might want to report it to them. But a simple workaround is to just round your cex values, e.g. use round(cex, 2) and 0.001 will become 0. 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.
Re: [R] Specifying strip.names in Lattice plots [UPDATE]
On Tue, 4 Mar 2014, Rich Shepard wrote: What I want to learn how to do is either 1) put the ylab in the strip or 2) eliminate the strip as redundant. The latest iteration, and the accompanying warning message: xyplot(cbind(dalles.disch.ts, dalles.temp.ts), main = Columbia River @ The Dalles, xlab = Date (Year.Month), ylab = c(Temperature (C),Discharge (cfs)), strip = FALSE, strip.left = strip.custom(par.strip.text = trellis.par.get(c(Mean Temperature,Mean Discharge Warning message: In if (name %in% names(lattice.theme[[.Device]])) lattice.theme[[.Device]][[name]] else NULL : the condition has length 1 and only the first element will be used Please correct the syntax in the xyplot() function so the two strips have the desired text. Thanks, Rich __ 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] Specifying strip.names in Lattice plots [SEMI-RESOLVED]
On Wed, 5 Mar 2014, Rich Shepard wrote: 2) eliminate the strip as redundant. Did this. It works. Would like to learn how to use par.strip.text to change the text on each strip, but removing them when redundant works for this figure. Rich __ 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] Shortest connected path in a matrix
Here is some example data (hopefully the monospace formatting is preserved): a b c d e - - - - - 1 | F | T | F | T | F | - - - - - 2 | T | F | T | F | T | - - - - - 3 | T | T | F | F | F | - - - - - 4 | F | T | F | T | F | - - - - - 5 | F | T | F | F | T | - - - - - So, for cell b1, the shortest possible path to a true value in row 5 is b1-a2-a3-b4-b5 (distance: sqrt(2) + 1 + sqrt(2) + 1). * Shortest paths are not necessarily unique, but I just need to find the length. * If it's computationally hard to guarantee the absolute shortest path, I can probably live with nearly shortest paths. * Paths can backtrack, so the shortest path from cell e2 to row 4 is e2-d1-c2-b3-b4-b5. I need to calculate the shortest path for all true cells to all rows further down the matrix. I'm afraid I'm going to have to write some sort of recursive path-tracing algorithm, but I'm hoping there's a package already in existence that accomplishes this already... -bryan On Tue, Mar 4, 2014 at 1:13 PM, McCloskey, Bryan bmcclos...@usgs.govwrote: I have a binary rectangular T/F matrix; I need to be able to calculate the shortest path (i.e., Pythagorean distance) between a populated cell in row j and any populated cell in some row j+n. For instance, if I have a chessboard with random black/white square colors, I need the shortest distance (linear distance, not number of steps) for a king to get from a specified black space on the first row, to _any_ black space in a specified further row, traveling only on black spaces. Any idea? Thanks, -bryan [[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] Shortest connected path in a matrix
(Mod my ignorance) This appears to be computer science/math problem and has nothing specifically to do with statistics nor R. So I suggest you post on a more appropriate venue rather than here. Cheers, Bert Bert Gunter Genentech Nonclinical Biostatistics (650) 467-7374 Data is not information. Information is not knowledge. And knowledge is certainly not wisdom. H. Gilbert Welch On Wed, Mar 5, 2014 at 9:44 AM, McCloskey, Bryan bmcclos...@usgs.govwrote: Here is some example data (hopefully the monospace formatting is preserved): a b c d e - - - - - 1 | F | T | F | T | F | - - - - - 2 | T | F | T | F | T | - - - - - 3 | T | T | F | F | F | - - - - - 4 | F | T | F | T | F | - - - - - 5 | F | T | F | F | T | - - - - - So, for cell b1, the shortest possible path to a true value in row 5 is b1-a2-a3-b4-b5 (distance: sqrt(2) + 1 + sqrt(2) + 1). * Shortest paths are not necessarily unique, but I just need to find the length. * If it's computationally hard to guarantee the absolute shortest path, I can probably live with nearly shortest paths. * Paths can backtrack, so the shortest path from cell e2 to row 4 is e2-d1-c2-b3-b4-b5. I need to calculate the shortest path for all true cells to all rows further down the matrix. I'm afraid I'm going to have to write some sort of recursive path-tracing algorithm, but I'm hoping there's a package already in existence that accomplishes this already... -bryan On Tue, Mar 4, 2014 at 1:13 PM, McCloskey, Bryan bmcclos...@usgs.gov wrote: I have a binary rectangular T/F matrix; I need to be able to calculate the shortest path (i.e., Pythagorean distance) between a populated cell in row j and any populated cell in some row j+n. For instance, if I have a chessboard with random black/white square colors, I need the shortest distance (linear distance, not number of steps) for a king to get from a specified black space on the first row, to _any_ black space in a specified further row, traveling only on black spaces. Any idea? Thanks, -bryan [[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.
Re: [R] Shortest connected path in a matrix
You may want to check bioconductor packages doing graph algorithms. Maybe this one: http://www.bioconductor.org/packages/release/bioc/manuals/RBGL/man/RBGL.pdf See for example ?dijkstra.sp On 5 March 2014 18:44, McCloskey, Bryan bmcclos...@usgs.gov wrote: Here is some example data (hopefully the monospace formatting is preserved): a b c d e - - - - - 1 | F | T | F | T | F | - - - - - 2 | T | F | T | F | T | - - - - - 3 | T | T | F | F | F | - - - - - 4 | F | T | F | T | F | - - - - - 5 | F | T | F | F | T | - - - - - So, for cell b1, the shortest possible path to a true value in row 5 is b1-a2-a3-b4-b5 (distance: sqrt(2) + 1 + sqrt(2) + 1). * Shortest paths are not necessarily unique, but I just need to find the length. * If it's computationally hard to guarantee the absolute shortest path, I can probably live with nearly shortest paths. * Paths can backtrack, so the shortest path from cell e2 to row 4 is e2-d1-c2-b3-b4-b5. I need to calculate the shortest path for all true cells to all rows further down the matrix. I'm afraid I'm going to have to write some sort of recursive path-tracing algorithm, but I'm hoping there's a package already in existence that accomplishes this already... -bryan On Tue, Mar 4, 2014 at 1:13 PM, McCloskey, Bryan bmcclos...@usgs.govwrote: I have a binary rectangular T/F matrix; I need to be able to calculate the shortest path (i.e., Pythagorean distance) between a populated cell in row j and any populated cell in some row j+n. For instance, if I have a chessboard with random black/white square colors, I need the shortest distance (linear distance, not number of steps) for a king to get from a specified black space on the first row, to _any_ black space in a specified further row, traveling only on black spaces. Any idea? Thanks, -bryan [[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] books and datasets
Dear R-users, I am looking for books and R examples that focus on the analysis of advertising, marketing, web metrics, and social media datasets. I wonder if you have recommendations for me. Thanks 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.
[R] Two maps with different legends with spplot
Hi all, I am using spplot to plot two maps in the same device. In my case, I have a map of a certain variable and the same map of the standard deviation of the same variable. The range of the second is much smaller than the range of the first so my wish is to plot both but to have them on different scales for the color legend (hence, to have two separated legends). Is it possible? I have worked out a way to do that, which passes through stretching the second variable to the range of the first, and then work hardly on the key list from within the trellis object structure. I was wondering if there is a more straightforward way. Thanks, Gianluca Inviato da iPhone __ 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] alternative to wireframe()
Have you looked at persp() ? -- Don MacQueen Lawrence Livermore National Laboratory 7000 East Ave., L-627 Livermore, CA 94550 925-423-1062 On 3/4/14 2:56 AM, Rainer M Krug rai...@krugs.de wrote: Hi I am slowly getting enough of wireframe() from the package lattice, as it is to complicated for what I need and does not really do what I want. I am using it to produce a (surprise!) wireframe plot (see example below). The plot function is part of plot.tss in a package which I am working on, and the plot looks nice, but I would like to have the option to use it as part of a layout(), which does not work. Also, I did not get a real handle if I have to wrap the wireframe function into a print(). So I am looking for a simpler alternative, which produces a similar plot to the one created by the example below. I would prefer a different legend and I would like to have the 3D plot respect the layout() command above. Any suggestions which function I could use? Thanks, Rainer --8---cut here---start-8--- library(lattice) x - structure(list(threshold1 = c(21, 74.5, 128, 181.5, 235, 21, 74.5, 128, 181.5, 235, 21, 74.5, 128, 181.5, 235, 21, 74.5, 128, 181.5, 235, 21, 74.5, 128, 181.5, 235), threshold2 = c(0, 0, 0, 0, 0, 61.25, 61.25, 61.25, 61.25, 61.25, 122.5, 122.5, 122.5, 122.5, 122.5, 183.75, 183.75, 183.75, 183.75, 183.75, 245, 245, 245, 245, 245), overallAccuracy = c(0.606076276664512, 0.606076276664512, 0.606076276664512, 0.606076276664512, 0.606076276664512, 0.594182288299935, 0.597026502908856, 0.605559146735617, 0.606076276664512, 0.606076276664512, 0.6120232708468, 0.62262443438914, 0.633742727860375, 0.611376858435682, 0.606076276664512, 0.618616677440207, 0.630122818358112, 0.706658047834518, 0.695151906916613, 0.606076276664512, 0.393923723335488, 0.405429864253394, 0.482740788623142, 0.58655462184874, 0.606076276664512), sensitivity = c(0, 0, 0, 0, 0, 0.012471283229406, 0.0118148999015425, 0.00393829996718083, 0, 0, 0.130620282244831, 0.126025598949787, 0.0912372825730226, 0.0141122415490647, 0, 0.598949786675418, 0.594355103380374, 0.526419428946505, 0.296685264194289, 0, 1, 0.995405316704956, 0.927469642271086, 0.661962586150312, 0), specificity = c(1, 1, 1, 1, 1, 0.972269624573379, 0.977389078498293, 0.996587030716723, 1, 1, 0.924914675767918, 0.945392491467577, 0.986348122866894, 0.99957337883959, 1, 0.631399317406143, 0.653370307167236, 0.823805460750853, 0.954138225255973, 1, 0, 0.0219709897610922, 0.193686006825939, 0.537542662116041, 1), tss = c(0, 0, 0, 0, 0, -0.0152590921972152, -0.010796021600164, 0.000525330683904368, 0, 0, 0.0555349580127491, 0.0714180904173634, 0.0775854054399168, 0.0136856203886551, 0, 0.230349104081562, 0.24772541054761, 0.350224889697358, 0.250823489450262, 0, 0, 0.0173763064660479, 0.121155649097025, 0.199505248266353, 0), kappa = c(0, 0, 0, 0, 0, -0.0182824920390604, -0.0129583064846108, 0.000635541910639162, 0, 0, 0.0639760122213153, 0.0828765513813943, 0.0918579943188013, 0.0165372451769894, 0, 0.223843122923358, 0.242311121014362, 0.363093208690974, 0.28206036774748, 0, 0, 0.0137627491763651, 0.100591615720438, 0.187264280971332, 0), pP = c(0L, 0L, 0L, 0L, 0L, 38L, 36L, 12L, 0L, 0L, 398L, 384L, 278L, 43L, 0L, 1825L, 1811L, 1604L, 904L, 0L, 3047L, 3033L, 2826L, 2017L, 0L), pA = c(0L, 0L, 0L, 0L, 0L, 130L, 106L, 16L, 0L, 0L, 352L, 256L, 64L, 2L, 0L, 1728L, 1625L, 826L, 215L, 0L, 4688L, 4585L, 3780L, 2168L, 0L), aA = c(4688L, 4688L, 4688L, 4688L, 4688L, 4558L, 4582L, 4672L, 4688L, 4688L, 4336L, 4432L, 4624L, 4686L, 4688L, 2960L, 3063L, 3862L, 4473L, 4688L, 0L, 103L, 908L, 2520L, 4688L), aP = c(3047L, 3047L, 3047L, 3047L, 3047L, 3009L, 3011L, 3035L, 3047L, 3047L, 2649L, 2663L, 2769L, 3004L, 3047L, 1222L, 1236L, 1443L, 2143L, 3047L, 0L, 14L, 221L, 1030L, 3047L), n = 7735L), .Names = c(threshold1, threshold2, overallAccuracy, sensitivity, specificity, tss, kappa, pP, pA, aA, aP, n), class = TSS, link = .Primitive(), largerPres1 = TRUE, largerPres2 = FALSE, threshold1 = c(21, 74.5, 128, 181.5, 235), threshold2 = c(0, 61.25, 122.5, 183.75, 245), dimension = 2) columns - c(tss, sensitivity, specificity) column.col - rainbow(length(columns)) xp - rep(x$threshold1, length(columns)) yp - rep(x$threshold2, length(columns)) zp - NULL for (i in 1:length(columns)) { zp - c(zp, x[[columns[i]]]) } grp - rep(columns, each=length(x$tss)) ### layout(matrix(1:4, ncol=2), c(1,2,3,4)) ### wireframe( zp ~ xp * yp, xlab = threshold 1, ylab = threshold 2, zlab = paste(columns, collapse=\n), groups = grp, par.settings = simpleTheme( alpha = 0.7, col = column.col, ), scales = list(arrows = FALSE), auto.key = TRUE ) --8---cut here---end---8--- -- Rainer M. Krug email: RMKrugatgmaildotcom __ R-help@r-project.org mailing list
Re: [R] problem with previous code
Hi Arun, Yes, that last command m1[indx2N] - m2[sort(indx1)] did exactly the trick, now the variable and their values are perfectly matched. Thanks a lot for your great help. Best, Elio On Wed, Mar 5, 2014 at 1:17 AM, arun smartpink...@yahoo.com wrote: Hi Elio, If you change the last line of the code: m1[indx2N] - m2[indx1] to m1[indx2N] - m2[sort(indx1)] sum(m1[rownames(m1)==p79,]) #[1] 191 sum(m2[rownames(m2)==p79,]) #[1] 191 The problem is in the order of the colnames/rownames in each of the datasets. I tested it on something like: m1 - matrix(0,12,12,dimnames=rep(list(paste0(a,1:12)),2)) set.seed(49) m2 - matrix(sample(0:2,4*4,replace=TRUE),ncol=4,dimnames=rep(list(c(a4,a6,a9,a12)),2)) which didn't show the problem. But, if I change the order of colnames: m1 - matrix(0,12,12,dimnames=rep(list(paste0(a,c(1:3,8:12,4:7))),2)) set.seed(49) m2 - matrix(sample(0:2,4*4,replace=TRUE),ncol=4,dimnames=rep(list(c(a9,a4,a6,a12)),2)) vec1 - paste0(rownames(m1)[row(m1)],colnames(m1)[col(m1)]) vec2 - paste0(rownames(m2)[row(m2)],colnames(m2)[col(m2)]) indx - match(vec1,vec2) indx1 - indx[!is.na(indx)] indx2 - match(vec2,vec1) indx2N - indx2[!is.na(indx2)] m1[indx2N] - m2[sort(indx1)] m1 Hope this helps. A.K. On Tuesday, March 4, 2014 3:16 PM, Elio Shijaku sel...@gmail.com wrote: Hi Arun, Sorry to disturb, but while trying your code to combine two matrices of unequal dimensions, I noticed a problem, the sum for each variable does not match, but the overall variables sum does match for both matrices, something isn't going right, perhaps the variables are getting mixed up, any idea?? Here is the code you sent me and the zipped files for testing: dat1 - read.table(mtest.txt,header=TRUE) dim(dat1) dat2 - read.table(1998res_x.txt,header=TRUE) dim(dat2) m1 - as.matrix(dat1) m2 - as.matrix(dat2) vec1 - paste0(rownames(m1)[row(m1)],colnames(m1)[col(m1)]) vec2 - paste0(rownames(m2)[row(m2)],colnames(m2)[col(m2)]) indx - match(vec1,vec2) indx1 - indx[!is.na(indx)] indx2 - match(vec2,vec1) indx2N - indx2[!is.na(indx2)] m1[indx2N] - m2[indx1] Thanks a lot!! Best, Elio [[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] Script question
You haven't said what, exactly, doesn't execute properly means (please read the posting guide). At a guess, you need to put print() around the expressions whose output you aren't seeing. -Don -- Don MacQueen Lawrence Livermore National Laboratory 7000 East Ave., L-627 Livermore, CA 94550 925-423-1062 On 3/5/14 6:19 AM, Milton O Faison mfai...@vsu.edu wrote: I have a script that works fine if I copy the whole thing from from a text editor and paste it into R, but doesn't execute properly if the file is called by the source command. Any clues on how to fix this? I can post the script if necessary. o M. Omar Faison, PhD Director, Office of Sponsored Programs and Research Virginia State University 804-631-3330 http://twitter.com/omarfaison ³The information in this email and any attachments may be confidential and privileged. Access to this email by anyone other than the intended addressee is unauthorized. If you are not the intended recipient (or the employee or agent responsible for delivering this information to the intended recipient) please notify the sender by reply email and immediately delete this email and any copies from your computer and/or storage system. The sender does not authorize the use, distribution, disclosure or reproduction of this email (or any part of its contents) by anyone other than the intended recipient(s). No representation is made that this email and any attachments are free of viruses. Virus scanning is recommended and is the responsibility of the recipient.² [[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] books and datasets
Google is your friend! Cheers, Bert Bert Gunter Genentech Nonclinical Biostatistics (650) 467-7374 Data is not information. Information is not knowledge. And knowledge is certainly not wisdom. H. Gilbert Welch On Wed, Mar 5, 2014 at 10:40 AM, Carlos Hernandez carlos.u...@gmail.comwrote: Dear R-users, I am looking for books and R examples that focus on the analysis of advertising, marketing, web metrics, and social media datasets. I wonder if you have recommendations for me. Thanks 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. [[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] histograms embedded in a plot (as alternative to jitter)
Not histograms, but here are two alternatives. The first gives you kernel density plots for each value and the second uses violin plots. Both plot points if there are fewer than 5. set.seed(42) y-rpois(500,2) x-rnorm(500,y,1) plot(x,y, type=n) for (i in seq(min(y), max(y), by=1)) { if (length(x[y==i])5) { points(x[y==i], rep(i, length(x[y==i]))) } else { a - density(x[y==i]) polygon(a$x, a$y+i, col=gray) } } require(vioplot) plot(x,y, type=n) for (i in seq(min(y), max(y), by=1)) { if (length(x[y==i])5) { points(x[y==i], rep(i, length(x[y==i]))) } else { vioplot(x[y==i], horizontal=TRUE, add=TRUE, at=i, col=gray) } } - David L Carlson Department of Anthropology Texas AM University College Station, TX 77840-4352 -Original Message- From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On Behalf Of Bernard North Sent: Wednesday, March 5, 2014 7:44 AM To: r-help@r-project.org Subject: [R] histograms embedded in a plot (as alternative to jitter) Dear R list, I am plotting a discrete valued number on the y axis against a continuous variable on the x axis. To allow sample size to be viewed for the discrete groups I am using vertical jitter. So my code is along the lines of y-rpois(500,2) x-rnorm(500,y,1) plot(x,jitter(y)) It has not been suggested that a more informative view of the sample size might be if a histogram could be inserted into the plot (instead of the jittered rows of points) for the values at each y-value Many thanks if anyone can think of a way to do this This email may contain information that is privileged, confidential or otherwise protected from disclosure. It must not be used by, or its contents copied or disclosed to, persons other than the addressee. If you have received this email in error please notify the sender immediately and delete the email. __ 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] Specifying strip.names in Lattice plots [UPDATE]
Hi Rich Without an example it is hard to work out. trellis.par.get(c(... is not correct, see names(trellis.par.get()) try strip= strip.custom(factor.levels = c(string vector of names for each panel), par.strip.text = list(cex = 0.8) ), Duncan Duncan Mackay Department of Agronomy and Soil Science University of New England Armidale NSW 2351 Email: home: mac...@northnet.com.au -Original Message- From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On Behalf Of Rich Shepard Sent: Thursday, 6 March 2014 03:29 To: r-help@r-project.org Subject: Re: [R] Specifying strip.names in Lattice plots [UPDATE] On Tue, 4 Mar 2014, Rich Shepard wrote: What I want to learn how to do is either 1) put the ylab in the strip or 2) eliminate the strip as redundant. The latest iteration, and the accompanying warning message: xyplot(cbind(dalles.disch.ts, dalles.temp.ts), main = Columbia River @ The Dalles, xlab = Date (Year.Month), ylab = c(Temperature (C),Discharge (cfs)), strip = FALSE, strip.left = strip.custom(par.strip.text = trellis.par.get(c(Mean Temperature,Mean Discharge Warning message: In if (name %in% names(lattice.theme[[.Device]])) lattice.theme[[.Device]][[name]] else NULL : the condition has length 1 and only the first element will be used Please correct the syntax in the xyplot() function so the two strips have the desired text. Thanks, Rich __ 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] Remove lattice panel
Dear all, Is there a possibility to remove the panel (with names) from a lattice plot? Thank you very much for your help! --- Catalin-Constantin ROIBU Lecturer PhD, Forestry engineer Forestry Faculty of Suceava Str. Universitatii no. 13, Suceava, 720229, Romania office phone +4 0230 52 29 78, ext. 531 mobile phone +4 0745 53 18 01 +4 0766 71 76 58 FAX:+4 0230 52 16 64 silvic.usv.ro [[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] AIX 7.1 and R build problems
Has anyone managed to build R-3.0.2 from source on AIX 7.1 using gcc 4.2.0. The configure script finishes with: ... checking whether wctrans exists and is declared... no checking whether iswblank exists and is declared... no checking whether wctype exists and is declared... no checking whether iswctype exists and is declared... no configure: error: Support for MBCS locales is required. Scanning through the config.log the configure script seems to be happy with the C99 compliance at hand. I have tried reading NEWS, README, and R-Help archives but I can't get past this. I'm all Google'd out. Is my compiler too old? _ Michael Beddo Senior Scientist Data Ventures, Inc. 1475 Central Ave. Suite 230 | Los Alamos, NM 87544 tel 505.695.2132 http://www.dataventures.com | Advanced - Effective - Actionable - Proven. Analytics __ 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] Remove lattice panel
I am not sure what you mean by remove (reproducible example??, code??) but see the skip argument of xyplot, which says: skip: A logical vector (default FALSE), replicated to be as long as the number of panels (spanning all pages). For elements that are TRUE, the corresponding panel position is skipped; i.e., nothing is plotted in that position. The panel that was supposed to be drawn there is now drawn in the next available panel position, and the positions of all the subsequent panels are bumped up accordingly. This may be useful for arranging plots in an informative manner. Presumably you can omit data that you don't wish to be plotted via the subset argument, or changing it to NA, or removing it from your data. If your issue is getting rid of unwanted empty factor levels, that can be done by the drop.unused.levels argument. If I haven't guessed what you mean, either explain more clearly or hope that someone else does. Cheers, Bert Bert Gunter Genentech Nonclinical Biostatistics (650) 467-7374 Data is not information. Information is not knowledge. And knowledge is certainly not wisdom. H. Gilbert Welch On Wed, Mar 5, 2014 at 1:11 PM, catalin roibu catalinro...@gmail.com wrote: Dear all, Is there a possibility to remove the panel (with names) from a lattice plot? Thank you very much for your help! --- Catalin-Constantin ROIBU Lecturer PhD, Forestry engineer Forestry Faculty of Suceava Str. Universitatii no. 13, Suceava, 720229, Romania office phone +4 0230 52 29 78, ext. 531 mobile phone +4 0745 53 18 01 +4 0766 71 76 58 FAX:+4 0230 52 16 64 silvic.usv.ro [[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] Remove lattice panel
Hi If you want to remove a panel with no plots in it see ?lattice::xyplot and look for drop.unused.levels If you want to remove a factor that has plot values in it then the easiest way may be to create a column of the factor and NA the values of the panel, relevel the factor and use the column instead Otherwise an example via dput would be needed for further help Duncan Duncan Mackay Department of Agronomy and Soil Science University of New England Armidale NSW 2351 Email: home: mac...@northnet.com.au -Original Message- From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On Behalf Of catalin roibu Sent: Thursday, 6 March 2014 07:11 To: r-help@r-project.org Subject: [R] Remove lattice panel Dear all, Is there a possibility to remove the panel (with names) from a lattice plot? Thank you very much for your help! --- Catalin-Constantin ROIBU Lecturer PhD, Forestry engineer Forestry Faculty of Suceava Str. Universitatii no. 13, Suceava, 720229, Romania office phone +4 0230 52 29 78, ext. 531 mobile phone +4 0745 53 18 01 +4 0766 71 76 58 FAX:+4 0230 52 16 64 silvic.usv.ro [[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] histograms embedded in a plot (as alternative to jitter)
On 03/06/2014 12:44 AM, Bernard North wrote: Dear R list, I am plotting a discrete valued number on the y axis against a continuous variable on the x axis. To allow sample size to be viewed for the discrete groups I am using vertical jitter. So my code is along the lines of y-rpois(500,2) x-rnorm(500,y,1) plot(x,jitter(y)) It has not been suggested that a more informative view of the sample size might be if a histogram could be inserted into the plot (instead of the jittered rows of points) for the values at each y-value Many thanks if anyone can think of a way to do this Hi Bernard, You can use the subplot function in the TeachingDemos package to do things like this. Jim __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] ZeligChoice - error
Dear R-users, nbsp; On the basis of ZeligChoice Manual I wrote several commands to create a plot of probabilities for a bivariate logit model. However, after sim() command I receive information about non-conformable arguments. Could you interpret the message? R is, as far as I know, the only statistical programme that enable to estimate not only bivariate probit, but also bivariate logit model... nbsp; Regards, Thank you in advance nbsp; library(ZeligChoice) danelt;-read.csv(dlaer.csv, header=TRUE, sep=;) z.out2lt;-zelig(list(mu1=c1~age+male+company+educ_4+educ_3+educ_2,mu2=c2~age+male+eduk_4+eduk_3+eduk_2),model=blogit,data=dane) x.low2 lt;- setx(z.out2, male = 1) x.high2 lt;- setx(z.out2, male = 0) s.out2 lt;- sim(z.out2, x1 = x.low2, x = x.high2, num = 500) AFTER THIS COMMAND I RECEIVE THE ANSWER ABOUT AN EROR: Error in all.coef[[1]] %*% as.matrix(xm[[1]]) : non-conformable arguments plot(s.out2) summary(z.out2) coef(z.out2) vcov(z.out2) nbsp; nbsp; nbsp; -- View this message in context: http://r.789695.n4.nabble.com/ZeligChoice-error-tp4686285.html Sent from the R help mailing list archive at Nabble.com. [[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] Fitting two parameter observations into copulas
Hi, I have one set of observations containing two parameters. How to fit it into copula (estimate the parameter of the copula and the margin function)? Let's say the margin distribution are log-normal distributions, and the copula is Gumbel copula. The data is as below: 1 974.0304 1010 2 6094.2672 1150 3 3103.2720 1490 4 1746.1872 1210 5 6683.7744 3060 6 6299.6832 3330 7 4784.0112 1550 8 1472.4288 607 9 3758.5728 1970 10 4381.2144 1350 11 5391.7056 2120 12 5353.3872 2560 13 3036.0960 2200 14 2016.9216 1190 15 2024.2224 670 16 2255.1264 970 17 2813.7024 1060 18 3889.5552 1290 19 2365.2950 2370 20 4471.6320 1450 21 2263.2048 1660 22 3047.9760 1170 23 2415.6144 2040 24 2249.5968 764 25 1379.7216 903 26 2743.3728 1140 27 2143.3680 1120 28 1396.9152 538 29 2385.5904 535 30 3966.5808 1720 31 3315.1680 1310 32 5822.1072 1660 33 2646.2160 1260 34 3592.9440 1050 35 6689.4336 1440 36 802.5264 785 37 1371.2544 595 38 1474.3728 390 39 1588.7232 731 40 5647.7952 1685 41 3832.0560 1245 42 2409.2208 1671 43 1978.6464 1073 44 6344.5248 1169 45 1654.9488 1380 46 3341.4336 2137 47 3563.0064 1543 48 4345.8336 825 49 1039.0464 572 50 727.4448 993 51 2271.0240 1360 52 5255.9712 1730 53 5972.0112 2150 54 2710.5840 1240 55 2314.3968 920 56 521.7264 572 57 3534.8832 1110 58 1471.4352 1010 59 2963.0448 1540 60 3431.1168 1320 61 2715.9408 1210 62 5161.7088 1260 63 2853.8352 870 64 4264.8336 1784 65 5605.5888 1720 66 3861.4752 1170 67 4537.5120 4290 68 4561.2720 2090 69 2652.7392 815 70 3610.1376 1670 71 6972.1776 2870 72 2654.4672 699 73 2982.0960 1020 74 3959.9280 1140 75 1224.2880 1370 76 2309.3860 2260 I have finished the univariate analysis for both parameter, estimated the distribution fitted the log-normal distribution. And I install the copula package, the function fitCopula required a parameter for the copula, I didn't quite understand, because this is the one which I am going to estimate. Also, how to convert the distribution (log-normal) to uniform marginal distribution for copula analysis? ibrary(copula) gumbel.cop - gumbelCopula(dim=2) myMvd - mvdc(gumbel.cop, c(lnorm,lnorm), list(list(meanlog = 7.1445391,sdlog=0.4568783), list(meanlog = 7.957392,sdlog=0.559831))) x - rmvdc(myMvd, 1000) fit - fitMvdc(x, myMvd, c(7.1445391,0.4568783,7.957392,0.559831)) the meanlog and sdlog value are derived from the data set. error message Error in if (alpha - 1 .Machine$double.eps^(1/3)) return(rCopula(n, : missing value where TRUE/FALSE needed How to choose the copula parameter with the given data, and the margin distributions derived from the data set? Thank you very much for your help. best regards, Yang Yang GFZ-Potsdam [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Is this a mistake in 'An Introduction to R'?
Peter I see there is no mistake. The phrase about the 'number of parameters' confused me, it is a little ambiguous. Many thanks for taking the time to help me. Geoff On 5 Mar 2014, at 11:20, Peter Dalgaard-2 [via R] ml-node+s789695n4686243...@n4.nabble.com wrote: On 04 Mar 2014, at 21:21 , Geoff Loveman [hidden email] wrote: In 'An Introduction to R', section 11.7 on nonlinear least squares fitting, the following example is given for obtaining the standard errors of the estimated parameters: To obtain the approximate standard errors (SE) of the estimates we do: sqrt(diag(2*out$minimum/(length(y) - 2) * solve(out$hessian)))The 2 in the line above represents the number of parameters. I know the inverted Hessian is multiplied by the mean square error and that the denominator of the MSE is the degrees of freedom (number of samples - number of parameters) but why does the numerator of the MSE (which is the RSS) get multiplied by the number of parameters? I have read through explanations of the method for obtaining the SE but I don't see where the MSE gets multiplied by the number of parameters or why this is needed as shown in the example? There are two 2's in that line, and I'd expect that only the last one has to do with the number of parameters, and the other one has to do with whether the Hessian is the second derivative of the sum of squares or of the negative loglikelihood function (half the sum of squares). Quick check: In a linear model, we have ssd = || Y- X beta ||^2 gradient = -2 (Y - X beta )'X Hessian H = 2 X'X and as we know, V(beta) = sigma^2 (X'X)^-1 = 2 sigma^2 H^-1 -pd Thanks for any help! Geoff Loveman Tech lead SMERAS QQ Maritime Life Support -- View this message in context: http://r.789695.n4.nabble.com/Is-this-a-mistake-in-An-Introduction-to-R-tp4686217.html Sent from the R help mailing list archive at Nabble.com. __ [hidden email] 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. -- Peter Dalgaard, Professor Center for Statistics, Copenhagen Business School Solbjerg Plads 3, 2000 Frederiksberg, Denmark Phone: (+45)38153501 Email: [hidden email] Priv: [hidden email] __ [hidden email] 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. If you reply to this email, your message will be added to the discussion below: http://r.789695.n4.nabble.com/Is-this-a-mistake-in-An-Introduction-to-R-tp4686217p4686243.html To unsubscribe from Is this a mistake in 'An Introduction to R'?, click here. NAML -- View this message in context: http://r.789695.n4.nabble.com/Is-this-a-mistake-in-An-Introduction-to-R-tp4686217p4686291.html Sent from the R help mailing list archive at Nabble.com. [[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] Building R for better performance
Simon, Thanks for the information and links. First of all, did you ever resolve your problem? If not, did you file an issue in Intel Premier Support? That's the best way to bring it to our attention. If you don't want to do that I can try to get a compiler or MKL support engineer to look at your Intel Developer Zone discussion. I have no experience with OS X, so I wouldn't be much help. I got the benchmark script, which I've attached, from Texas Advanced Computing Center. Here are my results (elapsed times, in secs): gcc build (default) icc/MKL build Creation, transp., deformation of a 5000x5000 matrix 3.25 2.95 5000x5000 normal distributed random matrix ^1000 5.13 1.52 Sorting of 14,000,000 random values 1.61 1.64 5600x5600 cross-product matrix (b = a' * a) 97.44 0.56 Linear regr. over a 4000x4000 matrix (c = a \ b') 46.06 0.49 FFT over 4,800,000 random values 0.65 0.61 Eigenvalues of a 1200x1200 random matrix 5.55 1.37 Determinant of a 5000x5000 random matrix 34.18 0.55 Cholesky decomposition of a 6000x6000 matrix 37.07 0.47 Inverse of a 3200x3200 random matrix 29.49 0.57 3,500,000 Fibonacci numbers calculation (vector calc) 1.310.38 Creation of a 6000x6000 Hilbert matrix (matrix calc) 0.77 0.99 Grand common divisors of 400,000 pairs (recursion) 0.63 0.56 Creation of a 1000x1000 Toeplitz matrix (loops) 2.24 2.34 Escoufier's method on a 90x90 matrix (mixed) 9.55 6.02 Total 274.93 21.01 Regards, Jonathan Anspach Sr. Software Engineer Intel Corp. jonathan.p.ansp...@intel.com 713-751-9460 -Original Message- From: Simon Zehnder [mailto:szehn...@uni-bonn.de] Sent: Wednesday, March 05, 2014 3:55 AM To: Anspach, Jonathan P Cc: r-help@r-project.org Subject: Re: [R] Building R for better performance Jonathan, I myself tried something like this - comparing gcc, clang and intel on a Mac. From my experiences in HPC on the university cluster (where we also use the Xeon Phi, Landeshochleistungscluster University RWTH Aachen), the Intel compiler has better code optimization in regard to vectorisation, etc. (clang is up to now suffering from a not yet implemented OpenMP library). Here is a revolutionanalytics article about this topic: http://blog.revolutionanalytics.com/2010/06/performance-benefits-of-multithreaded-r.html As I usually use the Rcpp package for C++ extensions this could give me further performance. Though, I already failed when trying to compile R with the Intel compiler and linking against the MKL (see my topic in the Intel developer zone: http://software.intel.com/en-us/comment/1767418 and my threads on the R-User list: https://stat.ethz.ch/pipermail/r-sig-mac/2013-November/010472.html). So, to your questions: 1) I think that most admins do not even use the Intel compiler to compile R - this seems to me rare. There are some people I know they do and I think they could be aware of it - but these are only a few. As R is growing in usage and I do know from regional user meetings that very large companies start using it in their BI units - this should be of interest. 2) I would really welcome this step because compilation with intel (especially on a Mac) and linking to the MKL seems to be delicate. I am interested in the data - so if it is possible send it via the list or directly to my account. Further, could you show some code that
Re: [R] Building R for better performance
Hi Jonathan, I think most people would be interested in such a tool, because main complaint of R is its slowness for some operations and big data. Even thought the intel software is paying , I could install it free since I am not selling any software and work for non-profit. I compiled successfully on my opensuse.. My question is : after make install , do I need to give special options to install.packages or they will be complied with icc automatically ? Regards CE -Original Message- From: Anspach, Jonathan P [jonathan.p.ansp...@intel.com] Date: 03/05/2014 12:28 AM To: r-help@r-project.org r-help@r-project.org Subject: [R] Building R for better performance Greetings, I'm a software engineer with Intel. Recently I've been investigating R performance on Intel Xeon and Xeon Phi processors and RH Linux. I've also compared the performance of R built with the Intel compilers and Intel Math Kernel Library to a default build (no config options) that uses the GNU compilers. To my dismay, I've found that the GNU build always runs on a single CPU core, even during matrix operations. The Intel build runs matrix operations on multiple cores, so it is much faster on those operations. Running the benchmark-2.5 on a 24 core Xeon system, the Intel build is 13x faster than the GNU build (21 seconds vs 275 seconds). Unfortunately, this advantage is not documented anywhere that I can see. Building with the Intel tools is very easy. Assuming the tools are installed in /opt/intel/composerxe, the process is simply (in bash shell): $ . /opt/intel/composerxe/bin/compilervars.sh intel64 $ ./configure --with-blas=-L/opt/intel/composerxe/mkl/lib/intel64 -lmkl_intel_lp64 -lmkl_intel_thread -lmkl_core -liomp5 -lpthread -lm --with-lapack CC=icc CFLAGS=-O2 CXX=icpc CXXFLAGS=-O2 F77=ifort FFLAGS=-O2 FC=ifort FCFLAGS=-O2 $ make $ make check My questions are: 1) Do most system admins and/or R installers know about this performance difference, and use the Intel tools to build R? 2) Can we add information on the advantage of building with the Intel tools, and how to do it, to the installation instructions and FAQ? I can post my data if anyone is interested. Thanks, Jonathan Anspach Sr. Software Engineer Intel Corp. jonathan.p.ansp...@intel.com 713-751-9460 __ 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] Survfit Error
Hi everyone, I am not new to R, but new to running survival models in R. I am trying to create some basic KM curves, using the following code: library(survival) library(KMsurv) (import data etc - basic right censored, with continuously observed time of death) sleepfit - survfit(Surv(timeb, death), data = sleep) Here timeb is measured is survival in years, death is a 1/0 indicator (1 = event = death), and the dataset is called sleep. But I get the following error Error in UseMethod(survfit, formula) : no applicable method for 'survfit' applied to an object of class Surv I'm not sure what I'm doing wrong, as I followed the example from the following paper http://anson.ucdavis.edu/~hiwang/teaching/10fall/R_tutorial%201.pdf , page 4 as a guide. I also tried sleepfit - survfit(Surv(timeb, death==1), data = sleep) and sleepfit - survfit(Surv(starttime, endtime, death), data = sleep) Can anyone tell me what I might be doing wrong? Thanks, Lucy [[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] AIX 7.1 and R build problems
Such questions belong on R-devel -- see the posting guide. On 05/03/2014 22:28, Mike Beddo wrote: Has anyone managed to build R-3.0.2 from source on AIX 7.1 using gcc 4.2.0. The configure script finishes with: ... checking whether wctrans exists and is declared... no checking whether iswblank exists and is declared... no checking whether wctype exists and is declared... no checking whether iswctype exists and is declared... no configure: error: Support for MBCS locales is required. Scanning through the config.log the configure script seems to be happy with the C99 compliance at hand. I have tried reading NEWS, README, and R-Help archives but I can't get past this. I'm all Google'd out. Is my compiler too old? No (but it is old and possible it was built incorrectly). Those functions are part of C99, and we have no idea why your compiler on your box is not finding them. As your IT support for help. _ Michael Beddo Senior Scientist Data Ventures, Inc. 1475 Central Ave. Suite 230 | Los Alamos, NM 87544 tel 505.695.2132 http://www.dataventures.com | Advanced - Effective - Actionable - Proven. Analytics __ 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, rip...@stats.ox.ac.uk 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.