Re: [R] How to plot two variables using a secondary Y axis
On 7/10/07, Felipe Carrillo [EMAIL PROTECTED] wrote: Date Fo Co6/27/2007 57.1 13.96/28/2007 57.7 14.3 6/29/2007 57.8 14.36/30/2007 57 13.97/1/2007 57.1 13.9 7/2/2007 57.2 14.07/3/2007 57.3 14.17/4/2007 57.6 14.2 7/5/2007 58 14.47/6/2007 58.1 14.57/7/2007 58.2 14.6 7/8/2007 58.4 14.77/9/200758.7 14.8 Hello all: I am a newbie to R, and I was wondering how can I plot the Temperature values above using Lattice or ggplot2 code. I want Date(X axis), Degrees F(Y axis) and Degrees C( on a secondary Y axis). Hi Felipe, It's not currently possible with ggplot2, but it is something on my to do list. Hadley __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Window Version for permtest
Hi Sir There is no Window Binary versin of package permtest.Please provide information in this regard. Thank you -- AMINA SHAHZADI Department of Statistics GC University Lahore, Pakistan. Email: [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Window Version for permtest
On Wed, 11 Jul 2007, amna khan wrote: Hi Sir There is no Window Binary versin of package permtest.Please provide information in this regard. See http://cran.r-project.org/bin/windows/contrib/2.5/check/permtest-check.log [Sending this once was quite sufficient.] -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Plot SpatialLinesDataFrame with xlim ylim
Michael, The plot method for SpatialLinesDataFrame objects resides in package sp, and questions regarding it are easier noticed on the r-sig-geo mailing list. The reason why they are plotted with aspect ratio 1 is that they are assumed to be spatial (geographical) data, and assume that 1 m north equals 1 m west -- think of a map. The exception is when the projection argument is set to longlat data (i.e. decimal degrees North/East), where the aspect ratio is computed differently, such that the argument above more or less holds. You should be able to override the default aspect setting by explicitly passing the e.g. asp=0.5 argument to plot. Here's the comment in the documentation of plot for Spatial objects (such as SpatialLinesDataFrame): The default aspect for map plots is 1; if however data are not projected (coordinates are longlat), the aspect is by default set to 1/cos(My * pi)/180) with My the y coordinate of the middle of the map (the mean of ylim, which defaults to the y range of bounding box). The argument |setParUsrBB| may be used to pass the logical value |TRUE| to functions within |plot.Spatial|. When set to |TRUE|, par(“usr”) will be overwritten with |c(xlim, ylim)|, which defaults to the bounding box of the spatial object. This is only needed in the particular context of graphic output to a specified device with given width and height, to be matched to the spatial object, when using par(“xaxs”) and par(“yaxs”) in addition to |par(mar=c(0,0,0,0))|. -- Edzer I'm running windows xp, R 2.3.1 with maptools 0.6-6, I guess. When plotting from a large SpatialLinesDataFrame and using xlim ylim to reduce the area, the plot axes automatically have the same scale size, even if xlim and ylim ranges differ. E.g.: tmp - readShapeLines(filepath) plot(tmp,xlim=c(-126,-119),ylim=c(50,51)) The y-axis range is actually 47-54, same range as the x-axis. What am I doing wrong? Should I be using a different object for simple coastline river data? Thanks in advance! Michael __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Improved Windows Vista compatibility
Duncan Murdoch and I have been working on Vista compatibility. The first build of R-2.5.1-win32.exe was built with a version of the installer that had some compatibility issues, and it has been rebuilt under the latest version. What now happens is that if you want to install R in an adminstrative area you need to select 'Run as Administrator' when running the installer: otherwise the default installation directory offered will be in your user area (e.g. c:/Users/ripley/Documents). The Tcl support files came with obsolete Winhelp 4 .hlp files that are not readable under Vista, and .chm help has been added. (As from the next release .hlp will be dropped.) Rtools.exe has also been rebuilt with newer versions of the tools that work on Vista. There is still a compiler path issue, and you need to add c:\Rtools\MinGW\libexec\gcc\mingw\3.4.5 (or similar) to your path. When that is done I have successfully built and run R-patched and R-devel on 64-bit Vista. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Installation on Leopard
Hi all, I'm wondering whether anyone has installed R on Mac OS X leopard with success. I downloaded the latest version, but couldn't have it installed. Any idea as to how I get around the problem? Thanks! Sincerely, Helin -- Helin (Colin) Gai Class of 2009, Duke University Box 96332 Durham, NC 27708 Phone: 919-943-6302 helin.gai at gmail.com hg9 at duke.edu [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Making Gehan-Breslow test for Survival data
Jose, The Gehan-Breslow test provides a generalization of the Kruskal-Wallis test for censored data. As an alternative, try using survdiff with rho=1. This method uses weights w(ti) = S(ti) (where S is the Kaplan-Meier estimate of survival) which yields Fleming and Harrington's version of the Kruskal-Wallis test for censored data. This test will give more weight to early differences in the hazards. Regards, -Cody Cody Hamilton Edwards Lifesciences Hi all, The survivals functions can be tested by the Log-rank test and others, for example the Gehan-Breslow. The graham breslow work with the alpha values. But I don't know how is the Gehan-Breslow test with R. Somebody know a type function?.. or other suggestions? Any help will be really appreciated José Bustos Marine Biologist Master Apllied Stat Program University of Concepción __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Previously saved workspace restored
hi there, i an beginner of R. some one have sent me a file (extension is .Rdata). i have installed R in my computer and i just double clicked the data. then it automatically opened R programme and displayed that [previously saved workspace restored]. the following message was displayed. Type 'demo()' for some demos, 'help()' for on-line help, or'help.start()' for an HTML browser interface to help.Type 'q()' to quit R. [Previously saved workspace restored] but how can I see the data (table) which is saved (in R format) in R?, i hope you will help me. Kristi Glover _ Explore the seven wonders of the world BRE [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] ECDF, distribution of Pareto, distribution of Normal
Thank you very much for your reply. I am afraid I have no idea what is wrong with the pgpg function. The parameters are generated from pre-fitted GPD distribution. 1.544 is the location parameter, 0.4373 is the scale parameter and -0.2398 is the shape parameter. Cound you please give me some hint? Stefan Grosse-2 wrote: Original Message Subject: [R] ECDF, distribution of Pareto, distribution of Normal From: livia [EMAIL PROTECTED] To: r-help@stat.math.ethz.ch Date: Tue Jul 10 2007 18:35:04 GMT+0200 Hello all, I would like to plot the emperical CDF, normal CDF and pareto CDF in the same graph and I amusing the following codes. z is a vector and I just need the part when z between 1.6 and 3. plot(ecdf(z), do.points=FALSE, verticals=TRUE, xlim=c(1.6,3),ylim=c(1-sum(z1.6)/length(z), 1)) x - seq(1.6, 3, 0.1) lines(x,pgpd(x, 1.544,0.4373,-0.2398), col=red) There is something wrong with your pgpd function, see ?pgpd for help and parameters... (I wonder how you got something plotted here...) y - seq(1.6, 3, 0.1) lines(y,pnorm(y, mean(z),sqrt(var(z))), col=blue) The emperical CDF and normal CDF look rather resonable, but the pareto CDF looks quite odd. I am not sure whether I plot the pareto CDF correctly e.g. in the right yaxs or any other mistake? At the same time, let t represents the vector whose values are larger than 1.6(the part we want). If I implement the following codes and plot the emperical CDF and pareto CDF, the pareto CDF seems fit. plot(ecdf(t), do.points=FALSE, verticals=TRUE) x - seq(1.6, 3, 0.1) lines(x,pgpd(x, 1.544,0.4373,-0.2398), col=red) Could anyone give me some advice on this? Many thanks. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- View this message in context: http://www.nabble.com/ECDF%2C-distribution-of-Pareto%2C-distribution-of-Normal-tf4056943.html#a11536305 Sent from the R help mailing list archive at Nabble.com. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] [R-sig-Geo] Plot SpatialLinesDataFrame with xlim ylim
On Wed, 11 Jul 2007, Edzer J. Pebesma wrote: Michael, The plot method for SpatialLinesDataFrame objects resides in package sp, and questions regarding it are easier noticed on the r-sig-geo mailing list. The reason why they are plotted with aspect ratio 1 is that they are assumed to be spatial (geographical) data, and assume that 1 m north equals 1 m west -- think of a map. The exception is when the projection argument is set to longlat data (i.e. decimal degrees North/East), where the aspect ratio is computed differently, such that the argument above more or less holds. You should be able to override the default aspect setting by explicitly passing the e.g. asp=0.5 argument to plot. Here's the comment in the documentation of plot for Spatial objects (such as SpatialLinesDataFrame): The default aspect for map plots is 1; if however data are not projected (coordinates are longlat), the aspect is by default set to 1/cos(My * pi)/180) with My the y coordinate of the middle of the map (the mean of ylim, which defaults to the y range of bounding box). The argument |setParUsrBB| may be used to pass the logical value |TRUE| to functions within |plot.Spatial|. When set to |TRUE|, par(?usr?) will be overwritten with |c(xlim, ylim)|, which defaults to the bounding box of the spatial object. This is only needed in the particular context of graphic output to a specified device with given width and height, to be matched to the spatial object, when using par(?xaxs?) and par(?yaxs?) in addition to |par(mar=c(0,0,0,0))|. Yes, if you look at how the GE_SpatialGrid() function in maptools works - you'll see how it meddles with the actual regional extents and the device size. I think Michael's data should also have been set to longlat: proj4string(tmp) - CRS(+proj=longlat) The key is realising that the axes are driven by the device shape, not by the xlim/ylim as such, as Edzer says. Roger -- Edzer I'm running windows xp, R 2.3.1 with maptools 0.6-6, I guess. When plotting from a large SpatialLinesDataFrame and using xlim ylim to reduce the area, the plot axes automatically have the same scale size, even if xlim and ylim ranges differ. E.g.: tmp - readShapeLines(filepath) plot(tmp,xlim=c(-126,-119),ylim=c(50,51)) The y-axis range is actually 47-54, same range as the x-axis. What am I doing wrong? Should I be using a different object for simple coastline river data? Thanks in advance! Michael ___ R-sig-Geo mailing list [EMAIL PROTECTED] https://stat.ethz.ch/mailman/listinfo/r-sig-geo -- Roger Bivand Economic Geography Section, Department of Economics, Norwegian School of Economics and Business Administration, Helleveien 30, N-5045 Bergen, Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43 e-mail: [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] type III ANOVA for a nested linear model
Indeed! And, apropos of the expression, to be Ripleyed (and so be condemned to eating cookies for a long, long time), what about being Billasted? BestR, Mark. Simon Blomberg-4 wrote: I second the nomination! Simon. On Tue, 2007-07-10 at 10:02 -0600, Greg Snow wrote: I nominate the following 2 pieces from Bill's reply for fortunes (probably 2 separate fortunes): All this becomes even more glaring if you take the unusal step of plotting the data. and What sort of editor would overlook this clear and demonstrable message leaping out from the data in favour of some arcane argument about types of sums of squares? Several answers come to mind: A power freak, a SAS afficianado, an idiot. -- Simon Blomberg, BSc (Hons), PhD, MAppStat. Lecturer and Consultant Statistician Faculty of Biological and Chemical Sciences The University of Queensland St. Lucia Queensland 4072 Australia Room 320 Goddard Building (8) T: +61 7 3365 2506 email: S.Blomberg1_at_uq.edu.au Policies: 1. I will NOT analyse your data for you. 2. Your deadline is your problem. The combination of some data and an aching desire for an answer does not ensure that a reasonable answer can be extracted from a given body of data. - John Tukey. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- View this message in context: http://www.nabble.com/type-III-ANOVA-for-a-nested-linear-model-tf4055192.html#a11536509 Sent from the R help mailing list archive at Nabble.com. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] question about gamm models
Dear R-Users, I have a question concerning mixed models in R. The strata of my model are the counties of Germany. The differencies between these counties should be modelled as realizations of a normally distributed random variable X. Moreover, the model contains a 0/1 variable A that enters as a fixed effect. The only special feature that should be additionally in the model is the following: In a usual mixed model the (constant) variance of X will be chosen in an optimal way, but I want to fit 2 constant variances, one for the subset {A=0} and the other one for the subset {A=1}. Nevertheless it should be one and the same random variable X. I know that it is possible to fit a model with two independent random variables X1 and X2 for the subsets {A=0} and {A=1} respectively. But I want it to be the same! Equivalently the correlation between X1 and X2 should be 1. Can anybody help me in this respect? Yours sincerely Matthias an der Heiden __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] variable and value labels
Hi Steve, I am new here - enjoying the power of R compared to SPSS. Looking for sets of tips and tricks for people with old SPSS habits. In particular, would like to know an easy way to set variable labels across a dataset and to set value labels for sets of variables. Maybe you should take a look at the following document : http://oit.utk.edu/scc/RforSASSPSSusers.pdf HTH, Julien -- Julien Barnier Groupe de recherche sur la socialisation ENS-LSH - Lyon, France __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Changing default library
Hallo R experts, I got a question concerning the .libPaths(). if I do the command .libPaths() than the result is /usr/lib/R/library. This is the default folder. I now want to change this one into /home/csc/usr/lib/R/library. I thought it would work with the command .libPaths(/home/csc/usr/lib/R/library). When I than do the command .libPaths() the result is: /usr/lib/R/library /home/csc/usr/lib/R/library. But if I start R the next time the result of the command .libPaths() is again just /usr/lib/R/library. Can anyone help me? Thanks, Corinna [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Changing default library
Hi, When I than do the command .libPaths() the result is: /usr/lib/R/library /home/csc/usr/lib/R/library. But if I start R the next time the result of the command .libPaths() is again just /usr/lib/R/library. From libPaths help page : , | The library search path is initialized at startup from the environment | variable R_LIBS (which should be a colon-separated list of directories | at which R library trees are rooted) by calling .libPaths with the | directories specified in R_LIBS. ` HTH, Julien -- Julien Barnier Groupe de recherche sur la socialisation ENS-LSH - Lyon, France __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Changing default library
Corinna: I got a question concerning the .libPaths(). if I do the command .libPaths() than the result is /usr/lib/R/library. This is the default folder. I now want to change this one into /home/csc/usr/lib/R/library. I thought it would work with the command .libPaths(/home/csc/usr/lib/R/library). When I than do the command .libPaths() the result is: /usr/lib/R/library /home/csc/usr/lib/R/library. But if I start R the next time the result of the command .libPaths() is again just /usr/lib/R/library. Can anyone help me? You should set the R_LIBS environment variable (that the man page of .libPaths points you to). See also FAQ 5.2. hth, Z __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] inquiry about anova and ancova
Dear R users, I have a rather knotty analysis problem and I was hoping that someone on this list would be able to help. I was advised to try this list by a colleague who uses R but it is a statistical inquiry not about how to use R. In brief I have a 3x2 anova, 2 tasks under 3 conditions, within subjects. I also took a variety of personality measures that might influence the results under the different conditions. I had thought that an ancova would be the best test, but it might be the case that this would not work with a within subjects design. I have not found anything that explicitly states whether or not it would, but all the examples I have read are between subjects design. I also thought of investigating a manova, but it is not really the case that I have more than one DV, it is the same DV in 6 different combinations of task and condition. There were 4 personality measures and I wanted to look at the degree to which they affected the task/ condition interaction. I have explained this briefly here, but I can of course provied more details to anyone who can advise me further with this. Thanks, Mary-Jane Anderson Information Analyst Platform Project Information Services Division, NHS National Services Scotland, Gyle Square, 1 South Gyle Crescent, Edinburgh, EH12 9EB. 0131 275 7163. _ NHS National Services Scotland Disclaimer The information contained in this message may be confidentia...{{dropped}} __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] variable and value labels
Dear list I am new here - enjoying the power of R compared to SPSS. Looking for sets of tips and tricks for people with old SPSS habits. In particular, would like to know an easy way to set variable labels across a dataset and to set value labels for sets of variables. Grateful for any help, Steve Powell __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] inquiry about anova and ancova
Anderson, Mary-Jane wrote: Dear R users, I have a rather knotty analysis problem and I was hoping that someone on this list would be able to help. I was advised to try this list by a colleague who uses R but it is a statistical inquiry not about how to use R. In brief I have a 3x2 anova, 2 tasks under 3 conditions, within subjects. I also took a variety of personality measures that might influence the results under the different conditions. I had thought that an ancova would be the best test, but it might be the case that this would not work with a within subjects design. I have not found anything that explicitly states whether or not it would, but all the examples I have read are between subjects design. I also thought of investigating a manova, but it is not really the case that I have more than one DV, it is the same DV in 6 different combinations of task and condition. There were 4 personality measures and I wanted to look at the degree to which they affected the task/ condition interaction. I have explained this briefly here, but I can of course provied more details to anyone who can advise me further with this. This sounds like a job for a Multivariate Linear Model (assuming that you have complete data for each subject or are prepared to throw away subjects with missing values). This lets you decompose the response into mean, effects of task and condition, and the interaction effect. Each component can then be separately tested for effect of predictors, using multivariate tests, or F tests under sphericity assumptions. Have a look at example(anova.mlm); this mostly looks at cases where effects are tested against zero, but the last example involves a (bogus) between subject factor f. -- O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] p-value from survreg, library(survival)
dear r experts: I would appreciate advice on how to get the p-value from the object 'sr' created with the function survreg() as given below. vlad sr-survreg(s~groups, dist=gaussian) Coefficients: (Intercept) groups -0.02138485 0.03868351 Scale= 0.01789372 Loglik(model)= 31.1 Loglik(intercept only)= 25.4 Chisq= 11.39 on 1 degrees of freedom, p= 0.00074 n= 16 [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] inquiry about anova and ancova
Hi Mary, it sounds like you have a split-plot design, or more gruesomely, a split-subject design. The model that I infer from your description of the design can be fit using the lme() function of the nlme() package, along the lines of a similar analysis documented in section 1.6 of Pinheiro and Bates (2000). It should also be possible by using the aov() function in base R. Cheers, Andrew On Wed, Jul 11, 2007 at 10:50:57AM +0100, Anderson, Mary-Jane wrote: Dear R users, I have a rather knotty analysis problem and I was hoping that someone on this list would be able to help. I was advised to try this list by a colleague who uses R but it is a statistical inquiry not about how to use R. In brief I have a 3x2 anova, 2 tasks under 3 conditions, within subjects. I also took a variety of personality measures that might influence the results under the different conditions. I had thought that an ancova would be the best test, but it might be the case that this would not work with a within subjects design. I have not found anything that explicitly states whether or not it would, but all the examples I have read are between subjects design. I also thought of investigating a manova, but it is not really the case that I have more than one DV, it is the same DV in 6 different combinations of task and condition. There were 4 personality measures and I wanted to look at the degree to which they affected the task/ condition interaction. I have explained this briefly here, but I can of course provied more details to anyone who can advise me further with this. Thanks, Mary-Jane Anderson Information Analyst Platform Project Information Services Division, NHS National Services Scotland, Gyle Square, 1 South Gyle Crescent, Edinburgh, EH12 9EB. 0131 275 7163. _ NHS National Services Scotland Disclaimer The information contained in this message may be confidentia...{{dropped}} __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Andrew Robinson Department of Mathematics and StatisticsTel: +61-3-8344-9763 University of Melbourne, VIC 3010 Australia Fax: +61-3-8344-4599 http://www.ms.unimelb.edu.au/~andrewpr http://blogs.mbs.edu/fishing-in-the-bay/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Warning message: cannot create HTML package index
On Fri, 6 Jul 2007, Peter Dalgaard wrote: Leo wrote: On 06/07/2007, Prof Brian Ripley wrote: On Fri, 6 Jul 2007, Leo wrote: I have set R_LIBS=~/R_lib as I don't have root access. The following message shown up every time after installing a package: .. The downloaded packages are in /tmp/RtmpBoIPoz/downloaded_packages Warning message: cannot create HTML package index in: tools:::unix.packages.html(.Library) Any ideas? It is a correct warning. What is the problem with being warned? R tries to maintain an HTML page of installed packages, but you don't have permission to update it. Where is that HTML page located on a GNU/Linux system? Is it possible to maintain a user HTML page of installed packages? Thanks, This confuses me a bit too. I had gotten used to the warning without thinking about it. It tries to update $RHOME/doc/html/packages.html, which starts like this: . ph3Packages in the standard library/h3 . However, if I run help.start, I get help.start() Making links in per-session dir ... If 'firefox' is already running, it is *not* restarted, and you must switch to its window. Otherwise, be patient ... and then it opens (say) file:///tmp/RtmpXyp5Cg/.R/doc/html/index.html which has a link to file:///tmp/RtmpXyp5Cg/.R/doc/html/packages.html which looks like this ph3Packages in /home/bs/pd/Rlibrary/h3 ph3Packages in /usr/lib64/R/library/h3 I.e. it is autogenerated by help.start and doesn't even look at the file in $RHOME. So what puzzles me is (a) why we maintain $RHOME/doc/html/packages.html at all One argument could be that this is browseable for everyone on a system, even without starting R. But then Some front-end GUIs use it (or used to use it). (b) why do we even try updating it when packages are installed in a private location? We don't know that. Given the use of symbolic links, it is not clear which libraries are private and which are links to .Library. However, we could be less cautious about this, and I've altered the code to update only if the path matches .Library exactly. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Installation on Leopard
On 7/11/07, Helin Gai [EMAIL PROTECTED] wrote: Hi all, I'm wondering whether anyone has installed R on Mac OS X leopard with success. I downloaded the latest version, but couldn't have it installed. Any idea as to how I get around the problem? Thanks! Simon Urbanek has comments about R for Leopard on his Wiki. Check R.research.att.com and wiki.Urbanek.info __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] p-value from survreg(), library(survival)
dear r experts: It seems my message got spam filtered, another try: i would appreciate advice on how to get the p-value from the object 'sr' created with the function survreg() as given below. vlad sr-survreg(s~groups, dist=gaussian) Coefficients: (Intercept) groups -0.02138485 0.03868351 Scale= 0.01789372 Loglik(model)= 31.1 Loglik(intercept only)= 25.4 Chisq= 11.39 on 1 degrees of freedom, p= 0.00074 n= 16 [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Some questions about quadratic programming (QP)
Dear R Users , As a beginner in QP, I'm trying to solve a Support Vector Machine problem by a QP. In particulare I am using the quadprog package. My questions are here: 1- In the document for the package (The quadprog Package), the inequality constraint is mentioned with = , however in a standard QP, this usaully is written with = . This constraint should be multiplied by a negativity sign to be standard ? 2- How about nonnegativity constraints ? Are that possible to be handled via upper and lower bounds? In document, nonnegativity is not mentioned. Is there any argument for that? 3- In QP formulation we have a constant 1/2 in objective function. This should be multiplied by the user or it is done by the algorithm? 4- More important, when I multiply a b*b matrix by D in quadratic term of objective function, I get thise message: Error in solve.QP(Dmat, dvec, Amat, bvec) : matrix D in quadratic function is not positive definite! How to multiply b*b matrix by the diagonal matrix ? Thank you very much for any help. Amir - Park yourself in front of a world of choices in alternative vehicles. [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] CDF for pareto distribution
Hi, I would like to use the following codes to plot the CDF for pareto distribution. Before doing this, I have plot the emperical one. x - seq(1.6, 3, 0.1) lines(x,pgpd(x, 1.544,0.4477557,), col=red) Could anyone give me some advice whether the above codes are correct? Many thanks. -- View this message in context: http://www.nabble.com/CDF-for-pareto-distribution-tf4061253.html#a11538272 Sent from the R help mailing list archive at Nabble.com. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Antigen found FILE FILTER= *.pif file
Antigen for Exchange found instruction.zip__instruction.doc .pif matching FILE FILTER= *.pif file filter. The file is currently Purged. The message, DELIVERY REPORTS ABOUT YOUR E_MAIL, was sent from [EMAIL PROTECTED] and was discovered in SMTP Messages\Inbound And Outbound located at University of Missouri/UM System/UM-TSMTPOUT1. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] RWeka control parameters classifiers interface
Hello, I have some trouble in achieving the desired parametrisation for the weka classifier functions, using the package RWeka. The problem is, that the functions result=classifier(formula, data, subset, na.action, control = Weka_control(mycontrol)) do not seem to be manipulated by the mycontrol- arguments Perhaps this should be resepected via the handlers- argument , but the documentation in this regard is rather sparse. # - Examples file =system.file(arff,iris.arff,package=RWeka) data =read.arff(file=file) rownames(data)=1:nrow(data) colnames(data)[ncol(data)] =class library(RWeka) # Example: no parameter influence mySMO =make_Weka_classifier(name=weka/classifiers/functions/SMO,class=NULL,handlers=list()); # Using control =Weka_control() m1 =mySMO(formula=class~.,data=data[,],control=Weka_control(K=weka.classifiers.functions.supportVector.PolyKernel,E=2)) m2 =mySMO(formula=class~.,data=data[,],control=Weka_control(K=weka.classifiers.functions.supportVector.PolyKernel,E=3)) m3 =mySMO(formula=class~.,data=data[,],control=c(K,weka.classifiers.functions.supportVector.PolyKernel,E,3)) # Using predefinded interface, does not work x1 =SMO(formula=class~.,data=data[,],control=Weka_control(K=weka.classifiers.functions.supportVector.PolyKernel,E=2)) x2 =SMO(formula=class~.,data=data[,],control=Weka_control(K=weka.classifiers.functions.supportVector.PolyKernel,E=3)) m1$call m2$call m3$call x1$call x2$call # no differences: m1 m2 m3 x1 x2 Any suggestions? Many thanks Bjoern many thanks bjoern __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Previously saved workspace restored
Hi If you enter the command ls()you will see a list of names that have come with the .Rdata file you double-clicked. If you enter one of these names at the command prompt you will see the data. So, for example if you have some data called mydata: ls() [1] mydata repos mydata [,1] [,2] [,3] [1,]147 [2,]258 [3,]369 Regards John --- -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Kristi Glover Sent: 11 July 2007 05:18 To: r-help@stat.math.ethz.ch Subject: [R] Previously saved workspace restored hi there, i an beginner of R. some one have sent me a file (extension is .Rdata). i have installed R in my computer and i just double clicked the data. then it automatically opened R programme and displayed that [previously saved workspace restored]. the following message was displayed. Type 'demo()' for some demos, 'help()' for on-line help, or'help.start()' for an HTML browser interface to help.Type 'q()' to quit R. [Previously saved workspace restored] but how can I see the data (table) which is saved (in R format) in R?, i hope you will help me. Kristi Glover _ Explore the seven wonders of the world BRE [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] type III ANOVA for a nested linear model
Hello Peter, thanks for your help. I'm quite sure that I specified the right model. Factor C is indeed nested within factor A. I think you were confused by the numbering of C (1..11), and it is easier to understand when I code it as you suggested (1,2,3 within each level of A, as in mydata1 [see below]). However, it does not matter which numbering I choose for carrying the analysis, as anova(lm(resp ~ A * B + (C %in% A), mydata)) anova(lm(resp ~ A * B + (C %in% A), mydata1)) both give the same results (as at least I had expected because of the nesting). However, I found that Anova() from the car package only accepts the second version. So, Anova(lm(resp ~ A * B + (C %in% A), mydata)) does not work (giving an error) but Anova(lm(resp ~ A * B + (C %in% A), mydata1)) does. This behaviour is rather confusing, or is there anything I'm missing? Thanks for your help again, Carsten R mydata A B C resp 1 1 1 1 34.12 2 1 1 2 32.45 3 1 1 3 44.55 4 1 2 1 20.88 5 1 2 2 22.32 6 1 2 3 27.71 7 2 1 6 38.20 8 2 1 7 31.62 9 2 1 8 38.71 102 2 6 18.93 112 2 7 20.57 122 2 8 31.55 133 1 9 40.81 143 1 10 42.23 153 1 11 41.26 163 2 9 28.41 173 2 10 24.07 183 2 11 21.16 R mydata1 A B C resp 1 1 1 1 34.12 2 1 1 2 32.45 3 1 1 3 44.55 4 1 2 1 20.88 5 1 2 2 22.32 6 1 2 3 27.71 7 2 1 1 38.20 8 2 1 2 31.62 9 2 1 3 38.71 102 2 1 18.93 112 2 2 20.57 122 2 3 31.55 133 1 1 40.81 143 1 2 42.23 153 1 3 41.26 163 2 1 28.41 173 2 2 24.07 183 2 3 21.16 On Tue, 2007-07-10 at 13:54 +0200, Peter Dalgaard wrote: Carsten Jaeger wrote: Hello, is it possible to obtain type III sums of squares for a nested model as in the following: lmod - lm(resp ~ A * B + (C %in% A), mydata)) I have tried library(car) Anova(lmod, type=III) but this gives me an error (and I also understand from the documentation of Anova as well as from a previous request (http://finzi.psych.upenn.edu/R/Rhelp02a/archive/64477.html) that it is not possible to specify nested models with car's Anova). anova(lmod) works, of course. My data (given below) is balanced so I expect the results to be similar for both type I and type III sums of squares. But are they *exactly* the same? The editor of the journal which I'm sending my manuscript to requests what he calls conventional type III tests and I'm not sure if can convince him to accept my type I analysis. In balanced designs, type I-IV SSD's are all identical. However, I don't think the model does what I think you think it does. Notice that nesting is used with two diferent meanings, in R it would be that the codings of C only makes sense within levels of A - e.g. if they were numbered 1:3 within each group, but with C==1 when A==1 having nothing to do with C==1 when A==2. SAS does something. er. else... What I think you want is a model where C is a random terms so that main effects of A can be tested, like in summary(aov(resp ~ A * B + Error(C), dd)) Error: C Df Sum Sq Mean Sq F value Pr(F) A 2 33.123 16.562 0.4981 0.6308 Residuals 6 199.501 33.250 Error: Within Df Sum Sq Mean Sq F value Pr(F) B 1 915.21 915.21 83.7846 9.57e-05 *** A:B2 16.138.07 0.7384 0.5168 Residuals 6 65.54 10.92 --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 (This is essentially the same structure as Martin Bleichner had earlier today, also @web.de. What is this? an epidemic? ;-)) __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] RWeka control parameters classifiers interface
On Wed, 11 Jul 2007 [EMAIL PROTECTED] wrote: The problem is, that the functions result=classifier(formula, data, subset, na.action, control = Weka_control(mycontrol)) do not seem to be manipulated by the mycontrol- arguments Yes, they are...not all parameter changes have always an effect on the specified learner. Perhaps this should be resepected via the handlers- argument , but the documentation in this regard is rather sparse. Handlers are not needed here. Re: sparse docs. In case you have not seen that paper already, there is a technical report on the ideas behind RWeka: http://epub.wu-wien.ac.at/dyn/openURL?id=oai:epub.wu-wien.ac.at:epub-wu-01_ba6 Re: SMO. Compare m1 - SMO(Species ~ ., data = iris) m2 - SMO(Species ~ ., data = iris, control = Weka_control( K = weka.classifiers.functions.supportVector.RBFKernel)) which yield different results so the Weka_control() works. The same happens if you register the mySMO() interface yourself. I'm not sure why the E = ... argument has no influence on the SMO, please check the Weka docs for this particular learner. Best, Z __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] variable and value labels
Steve Powell wrote: Dear list I am new here - enjoying the power of R compared to SPSS. Looking for sets of tips and tricks for people with old SPSS habits. In particular, would like to know an easy way to set variable labels across a dataset and to set value labels for sets of variables. Grateful for any help, Steve Powell In the Hmisc package see the functions spss.get, label, upData, and describe. Frank -- Frank E Harrell Jr Professor and Chair School of Medicine Department of Biostatistics Vanderbilt University __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] 3D plot and interactive PDFs
With version 8 of acrobat reader, it is now possible to have 3D in PDf documents. Does it exist already an R package who manage to produce 3D plots which can be saved as interactive 3D graphs in a PDF file? Best Regards Bruno Cavestro -- Leggi GRATIS le tue mail con il telefonino i-mode di Wind http://i-mode.wind.it/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] elementary statistics with R (rkward?)
Hi, I am trying to learn some basic statistics stuff but I cannot find any elementary statistics exercises using R language. Using RKward would be even better... I need that in analysing sociological data, obtained through questionnairres - findind corelations between variables, relations between different types of data, etc. Could anyone recommend simple tutorials/exercises, available on www for me to work on? I realize it would be much simple to do this introductory stuff with spss, that everyone around me is using here in Lithuania, but I'd really like to learn to do it with R instead... -- Donatas G. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] elementary statistics with R (rkward?)
I don't know anything about RKward, but there are many, many tutorials, guides and other documents written for people learning R available online. Try the introduction to R at: http://www.r-project.org/ under manuals, or some of the many fine contributions at: http://cran.r-project.org/other-docs.html Sarah On 7/11/07, Donatas G. [EMAIL PROTECTED] wrote: Hi, I am trying to learn some basic statistics stuff but I cannot find any elementary statistics exercises using R language. Using RKward would be even better... I need that in analysing sociological data, obtained through questionnairres - findind corelations between variables, relations between different types of data, etc. Could anyone recommend simple tutorials/exercises, available on www for me to work on? I realize it would be much simple to do this introductory stuff with spss, that everyone around me is using here in Lithuania, but I'd really like to learn to do it with R instead... -- Donatas G. -- Sarah Goslee http://www.functionaldiversity.org __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] elementary statistics with R (rkward?)
Face the music and buy the book: _Introductory Statistics with R_ by Peter Dalgaard. It's perfect for what you need. It's clear and concise and will teach you statistics AND R as painlessly as such a thing can be. It's inexpensive and you can get it on Amazon.com and every other major bookseller, including the nearest university bookstore. Charles Annis, P.E. [EMAIL PROTECTED] phone: 561-352-9699 eFax: 614-455-3265 http://www.StatisticalEngineering.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Donatas G. Sent: Wednesday, July 11, 2007 9:27 AM To: r-help@stat.math.ethz.ch Subject: [R] elementary statistics with R (rkward?) Hi, I am trying to learn some basic statistics stuff but I cannot find any elementary statistics exercises using R language. Using RKward would be even better... I need that in analysing sociological data, obtained through questionnairres - findind corelations between variables, relations between different types of data, etc. Could anyone recommend simple tutorials/exercises, available on www for me to work on? I realize it would be much simple to do this introductory stuff with spss, that everyone around me is using here in Lithuania, but I'd really like to learn to do it with R instead... -- Donatas G. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] CDF for pareto distribution
Le 07-07-11 à 07:56, livia a écrit : Hi, I would like to use the following codes to plot the CDF for pareto distribution. Before doing this, I have plot the emperical one. x - seq(1.6, 3, 0.1) lines(x,pgpd(x, 1.544,0.4477557,), col=red) Could anyone give me some advice whether the above codes are correct? Many thanks. livia, You seem to be struggling with the Pareto distribution... The above code seems correct, but you do not say where you took the pdpd() function from. This makes it harder for us to help you. In you other message (https://stat.ethz.ch/pipermail/r-help/2007-July/ 136137.html) you quote a negative scale parameter. The Pareto I know has strictly positive shape and scale parameters. Perhaps can you retry with functions ppareto() or pgenpareto() of package actuar. --- Vincent Goulet, Associate Professor École d'actuariat Université Laval, Québec [EMAIL PROTECTED] http://vgoulet.act.ulaval.ca __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Previously saved workspace restored
hi, thanks for the suggestion, however i am struggling to see the data, here is load(C://R//species.Rdata) ls()[1] abc abc1 abc2 example example1.14 generalislast.warningwhen i wrote ls(), then i saw the the files names I created before in R. there was no problem for the files which was created in R. but still i could not see the data which was received from email. species data was downloaded from email. i checked for .txt files. it worked but downloaded .Rdata is not working using ls(). is any supporting files needed to open the downloaded files? thanks for your suggestions bye Kristi Subject: RE: [R] Previously saved workspace restored Date: Wed, 11 Jul 2007 13:19:06 +0100 From: [EMAIL PROTECTED] To: [EMAIL PROTECTED]; r-help@stat.math.ethz.chHi If you enter the command ls()you will see a list of names that have come with the .Rdata file you double-clicked. If you enter one of these names at the command prompt you will see the data. So, for example if you have some data called mydata: ls() [1] mydata repos mydata [,1] [,2] [,3] [1,] 1 4 7 [2,] 2 5 8 [3,] 3 6 9 Regards John--- -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Kristi Glover Sent: 11 July 2007 05:18 To: r-help@stat.math.ethz.ch Subject: [R] Previously saved workspace restored hi there, i an beginner of R. some one have sent me a file (extension is .Rdata). i have installed R in my computer and i just double clicked the data. ! then it automatically opened R programme and displayed that [previously saved workspace restored]. the following message was displayed. Type 'demo()' for some demos, 'help()' for on-line help, or'help.start()' for an HTML browser interface to help.Type 'q()' to quit R. [Previously saved workspace restored] but how can I see the data (table) which is saved (in R format) in R?, i hope you will help me. Kristi Glover _ Explore the seven wonders of the world BRE [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. _ Discover the new Windows Vista [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] CDF for pareto distribution
Hi, thank you very much for your reply. The function pgpd() is from the package POT, and the 1.544 is the location parameter, 0.4477557 is the scale parameter and -0.50113 is the shape parameter, which can be both negtive or positive. Vincent Goulet wrote: Le 07-07-11 à 07:56, livia a écrit : Hi, I would like to use the following codes to plot the CDF for pareto distribution. Before doing this, I have plot the emperical one. x - seq(1.6, 3, 0.1) lines(x,pgpd(x, 1.544,0.4477557,-0.50113), col=red) Could anyone give me some advice whether the above codes are correct? Many thanks. livia, You seem to be struggling with the Pareto distribution... The above code seems correct, but you do not say where you took the pdpd() function from. This makes it harder for us to help you. In you other message (https://stat.ethz.ch/pipermail/r-help/2007-July/ 136137.html) you quote a negative scale parameter. The Pareto I know has strictly positive shape and scale parameters. Perhaps can you retry with functions ppareto() or pgenpareto() of package actuar. --- Vincent Goulet, Associate Professor École d'actuariat Université Laval, Québec [EMAIL PROTECTED] http://vgoulet.act.ulaval.ca __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- View this message in context: http://www.nabble.com/CDF-for-pareto-distribution-tf4061253.html#a11540928 Sent from the R help mailing list archive at Nabble.com. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] p-value from survreg(), library(survival)
str(survreg(s~groups, dist=gaussian)) is probably a good place to start. Hadley On 7/11/07, Vlado Sremac [EMAIL PROTECTED] wrote: dear r experts: It seems my message got spam filtered, another try: i would appreciate advice on how to get the p-value from the object 'sr' created with the function survreg() as given below. vlad sr-survreg(s~groups, dist=gaussian) Coefficients: (Intercept) groups -0.02138485 0.03868351 Scale= 0.01789372 Loglik(model)= 31.1 Loglik(intercept only)= 25.4 Chisq= 11.39 on 1 degrees of freedom, p= 0.00074 n= 16 [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Power calculation for the time series experiment
Hi All, We are planning to run an experiment, where samples will be taken at different time points (say, 0, 4, 8, 16, 24). If I am interested in the effect size of 1.5 for a reasonably large samples (say 500), what will be the power? Is it a good idea to use F-test (one-way ANOVA) as my test statistics? How can we include correlation structure among samples in the power analysis, if I use one-way ANOVA design? I am aware of power.anova.test() in R that will help me to do power calculation for one-way ANOVA. It will be of great help if you send me some related articles or pointers to some useful resources. Thanks in advance. Kind regards, Ezhil __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] 3D plot and interactive PDFs
On 7/11/2007 9:18 AM, Bruno C. wrote: With version 8 of acrobat reader, it is now possible to have 3D in PDf documents. Does it exist already an R package who manage to produce 3D plots which can be saved as interactive 3D graphs in a PDF file? No, not as far as I know. If you want to help to make it happen, I'd suggest working to add it to the GL2PS project (http://www.geuz.org/gl2ps) and then it should be automatically incorporated into the rgl package. Duncan Murdoch __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] type III ANOVA for a nested linear model
Billasted sounds too brutal. How about Billeted, as in what one does to one's breshly-caught bish? Andrew On Wed, Jul 11, 2007 at 01:48:10AM -0700, Mark Difford wrote: Indeed! And, apropos of the expression, to be Ripleyed (and so be condemned to eating cookies for a long, long time), what about being Billasted? BestR, Mark. Simon Blomberg-4 wrote: I second the nomination! Simon. On Tue, 2007-07-10 at 10:02 -0600, Greg Snow wrote: I nominate the following 2 pieces from Bill's reply for fortunes (probably 2 separate fortunes): All this becomes even more glaring if you take the unusal step of plotting the data. and What sort of editor would overlook this clear and demonstrable message leaping out from the data in favour of some arcane argument about types of sums of squares? Several answers come to mind: A power freak, a SAS afficianado, an idiot. -- Simon Blomberg, BSc (Hons), PhD, MAppStat. Lecturer and Consultant Statistician Faculty of Biological and Chemical Sciences The University of Queensland St. Lucia Queensland 4072 Australia Room 320 Goddard Building (8) T: +61 7 3365 2506 email: S.Blomberg1_at_uq.edu.au Policies: 1. I will NOT analyse your data for you. 2. Your deadline is your problem. The combination of some data and an aching desire for an answer does not ensure that a reasonable answer can be extracted from a given body of data. - John Tukey. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- View this message in context: http://www.nabble.com/type-III-ANOVA-for-a-nested-linear-model-tf4055192.html#a11536509 Sent from the R help mailing list archive at Nabble.com. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Andrew Robinson Department of Mathematics and StatisticsTel: +61-3-8344-9763 University of Melbourne, VIC 3010 Australia Fax: +61-3-8344-4599 http://www.ms.unimelb.edu.au/~andrewpr http://blogs.mbs.edu/fishing-in-the-bay/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] ECDF, distribution of Pareto, distribution of Normal
livia wrote: Hello all, I would like to plot the emperical CDF, normal CDF and pareto CDF in the same graph and I amusing the following codes. z is a vector and I just need the part when z between 1.6 and 3. plot(ecdf(z), do.points=FALSE, verticals=TRUE, xlim=c(1.6,3),ylim=c(1-sum(z1.6)/length(z), 1)) x - seq(1.6, 3, 0.1) lines(x,pgpd(x, 1.544,0.4373,-0.2398), col=red) y - seq(1.6, 3, 0.1) lines(y,pnorm(y, mean(z),sqrt(var(z))), col=blue) The emperical CDF and normal CDF look rather resonable, but the pareto CDF looks quite odd. I am not sure whether I plot the pareto CDF correctly e.g. in the right yaxs or any other mistake? At the same time, let t represents the vector whose values are larger than 1.6(the part we want). If I implement the following codes and plot the emperical CDF and pareto CDF, the pareto CDF seems fit. plot(ecdf(t), do.points=FALSE, verticals=TRUE) x - seq(1.6, 3, 0.1) lines(x,pgpd(x, 1.544,0.4373,-0.2398), col=red) Could anyone give me some advice on this? Many thanks. If any of your data points are less than 1.6, ecdf(z) and ecdf(t) will be different functions: for arguments greater than 1.6, the former will take values in c(mean(z1.6),1) and the latter will cover the range (0,1). It is not surprising that your pgpd function will fit only one of these empirical cdf's closely. Assuming that those GPD parameters were obtained by fitting to just the data values greater than 1.6, the GPD curve in your first plot should be u-mean(z1.6) x-seq(1.6,3,0.1) lines(x, u + (1-u)*pgpd(x, parameters ) J. R. M. Hosking __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] p-value from survreg(), library(survival)
Try also: pchisq(summary(sr)$chi, degrees_freedom, lower=FALSE) *You need know your degrees of freedom -- Henrique Dallazuanna Curitiba-Paraná-Brasil 25° 25' 40 S 49° 16' 22 O On 11/07/07, Vlado Sremac [EMAIL PROTECTED] wrote: dear r experts: It seems my message got spam filtered, another try: i would appreciate advice on how to get the p-value from the object 'sr' created with the function survreg() as given below. vlad sr-survreg(s~groups, dist=gaussian) Coefficients: (Intercept) groups -0.02138485 0.03868351 Scale= 0.01789372 Loglik(model)= 31.1 Loglik(intercept only)= 25.4 Chisq= 11.39 on 1 degrees of freedom, p= 0.00074 n= 16 [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] survfit for interval censored data
pHello,/ppI am a new R-user and would like to use survfit for interval censored data. Whenever I try I get an error message that states I can only use survfit for right censored or counting process data. I was wondering if anyone knows if there is an additional package available that can calculate KM curves for interval censored data, or another program with this capability?/ppThank you./ppS. Ellis (student)/p __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] p-value from survreg(), library(survival)
Actually, in this case, looking at the code for: survival:::print.survreg would be better, as the p value is calculate there, rather than being part of the survreg object. As with many R functions, the p value is calculated in the print method for the object. In this case, it is a pretty straightforward p value for the chi-square statistic. Using the output of Vlado's example below: format(signif(1 - pchisq(11.39, 1), 2)) [1] 0.00074 HTH, Marc Schwartz On Wed, 2007-07-11 at 16:11 +0200, hadley wickham wrote: str(survreg(s~groups, dist=gaussian)) is probably a good place to start. Hadley On 7/11/07, Vlado Sremac [EMAIL PROTECTED] wrote: dear r experts: It seems my message got spam filtered, another try: i would appreciate advice on how to get the p-value from the object 'sr' created with the function survreg() as given below. vlad sr-survreg(s~groups, dist=gaussian) Coefficients: (Intercept) groups -0.02138485 0.03868351 Scale= 0.01789372 Loglik(model)= 31.1 Loglik(intercept only)= 25.4 Chisq= 11.39 on 1 degrees of freedom, p= 0.00074 n= 16 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] p-value from survreg(), library(survival)
On 7/11/07, Marc Schwartz [EMAIL PROTECTED] wrote: Actually, in this case, looking at the code for: survival:::print.survreg would be better, as the p value is calculate there, rather than being part of the survreg object. As with many R functions, the p value is calculated in the print method for the object. I wish print methods wouldn't do that. Printing is supposed to be about displaying existing create, not creating new values. Hadley __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] CDF for pareto distribution
Le 07-07-11 à 10:01, livia a écrit : Hi, thank you very much for your reply. The function pgpd() is from the package POT, and the 1.544 is the location parameter, 0.4477557 is the scale parameter and -0.50113 is the shape parameter, which can be both negtive or positive. The documentation of POT states that the scale parameter must be positive. You seem right about the shape parameter, though; POT uses a different parametrization from the one I'm used to. The problem with x - seq(1.6, 3, 0.1) lines(x,pgpd(x, 1.544,0.4477557,-0.50113), col=red) is that lines() must be used to add to an already existing graphic. For a new plot, use x - seq(1.6, 3, 0.1) plot(x,pgpd(x, 1.544,0.4477557,-0.50113), col=red) which works just fine here. We cannot help you more if you do not give a *reproducible* example of what is going wrong. If you have issues with the pgpd() function, you should contact the maintainer of package POT, as the Posting Guide asks. HTH Vincent Goulet wrote: Le 07-07-11 à 07:56, livia a écrit : Hi, I would like to use the following codes to plot the CDF for pareto distribution. Before doing this, I have plot the emperical one. x - seq(1.6, 3, 0.1) lines(x,pgpd(x, 1.544,0.4477557,-0.50113), col=red) Could anyone give me some advice whether the above codes are correct? Many thanks. livia, You seem to be struggling with the Pareto distribution... The above code seems correct, but you do not say where you took the pdpd() function from. This makes it harder for us to help you. In you other message (https://stat.ethz.ch/pipermail/r-help/2007- July/ 136137.html) you quote a negative scale parameter. The Pareto I know has strictly positive shape and scale parameters. Perhaps can you retry with functions ppareto() or pgenpareto() of package actuar. --- Vincent Goulet, Associate Professor École d'actuariat Université Laval, Québec [EMAIL PROTECTED] http://vgoulet.act.ulaval.ca __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- View this message in context: http://www.nabble.com/CDF-for-pareto- distribution-tf4061253.html#a11540928 Sent from the R help mailing list archive at Nabble.com. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting- guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] 3D plot and interactive PDFs
This does not answer exactly to your question, anyway..: If you are planning to use latex, the package movie15 allows to include media files in your document (to be processed via pdflatex) vito Bruno C. wrote: With version 8 of acrobat reader, it is now possible to have 3D in PDf documents. Does it exist already an R package who manage to produce 3D plots which can be saved as interactive 3D graphs in a PDF file? Best Regards Bruno Cavestro -- Leggi GRATIS le tue mail con il telefonino i-mode™ di Wind http://i-mode.wind.it/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Vito M.R. Muggeo Dip.to Sc Statist e Matem `Vianelli' Università di Palermo viale delle Scienze, edificio 13 90128 Palermo - ITALY tel: 091 6626240 fax: 091 485726/485612 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] R CMD SHLIB problem with -lg2c and make:
Hello, I have 2 problems with R CMD SHLIB. The first problem didn't resolve, it just morphed into the new problem. I first have to admit that even though I've used R CMD SHLIB for a few years, I don't understand C compiling. The problem arose after our linux guy upgraded from Mandriva 2005 to Mandriva 2007. I was using R 2.4.1, but now it's upgraded to the current version. I used R CMD SHLIB to compile a C file that also linked to some fortran code. The error was something along the lines of not being able to find -lg2c. The gcc version I'm using is now 4.1.1, so our linux guy suggested the problem might be related to gfortran and f2c. I'm not sure... is this enough information to know what might be going on? So, I started looking around the FAQs online to try and figure out the problem and stumbled upon the R CMD COMPILE command. I ran it on my C and fortran code and didn't see any errors. I thought that was funny, so I erased the .o files it had created and tried again, with COMPILE and SHLIB... but now I get the error make: *** No rule to make target 'mhroc.o'. Stop. I have no idea what I did or how to correct it. Can anyone help? Thank you, Richard Zur __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] installing, removing, upgrading, and downgrading packages
I'm very new to R, trying to learn it. I started with R 2.4, but I have since upgraded to 2.5.0, on WindowsXP. I understand that 2.5.1 is now available. Last night in the course of things I loaded libraries coin and survival. I received warning messages that they had been built under version 2.5.1. However, as far as I could tell, they worked OK. But this brought some questions to mind: I found the remove.packages() command. Is there a way to revert a package back to a previous version without removing it? If not, and I remove one version, how do I specify downloading a particular (earlier) version? Was that warning of any consequence? Are there times when a package will not work properly with an earlier version of R, and will this be obvious to me when I try to use it? Or in a more general sense, are there foreseeable circumstances in which an older version of a package would be necessary, rather than the newest one? Should I keep my packages in a library folder outside of the R install folder? Right now, it appears that all the packages I download get installed into a library subdirectory under my R250 directory. Advantages and disadvantages of either method? Thanks. -- Christopher W. Ryan, MD SUNY Upstate Medical University Clinical Campus at Binghamton 40 Arch Street, Johnson City, NY 13790 cryanatbinghamtondotedu PGP public keys available at http://home.stny.rr.com/ryancw/ If you want to build a ship, don't drum up the men to gather wood, divide the work and give orders. Instead, teach them to yearn for the vast and endless sea. [Antoine de St. Exupery] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] elementary statistics with R (rkward?)
As a fellow beginner, I also found Handbook of Statistical Analyses Using R, by Brian Everitt, to be a very useful book. There is an accompanying R package, HSAUR. Also Using R for Introductory Statistics, by John Verzani. There is an accompanying R package, UsingR. Christopher W. Ryan, MD SUNY Upstate Medical University Clinical Campus at Binghamton 40 Arch Street, Johnson City, NY 13790 cryanatbinghamtondotedu PGP public keys available at http://home.stny.rr.com/ryancw/ If you want to build a ship, don't drum up the men to gather wood, divide the work and give orders. Instead, teach them to yearn for the vast and endless sea. [Antoine de St. Exupery] Charles Annis, P.E. wrote: Face the music and buy the book: _Introductory Statistics with R_ by Peter Dalgaard. It's perfect for what you need. It's clear and concise and will teach you statistics AND R as painlessly as such a thing can be. It's inexpensive and you can get it on Amazon.com and every other major bookseller, including the nearest university bookstore. Charles Annis, P.E. [EMAIL PROTECTED] phone: 561-352-9699 eFax: 614-455-3265 http://www.StatisticalEngineering.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Donatas G. Sent: Wednesday, July 11, 2007 9:27 AM To: r-help@stat.math.ethz.ch Subject: [R] elementary statistics with R (rkward?) Hi, I am trying to learn some basic statistics stuff but I cannot find any elementary statistics exercises using R language. Using RKward would be even better... I need that in analysing sociological data, obtained through questionnairres - findind corelations between variables, relations between different types of data, etc. Could anyone recommend simple tutorials/exercises, available on www for me to work on? I realize it would be much simple to do this introductory stuff with spss, that everyone around me is using here in Lithuania, but I'd really like to learn to do it with R instead... __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] type III ANOVA for a nested linear model
Carsten Jaeger wrote: Hello Peter, thanks for your help. I'm quite sure that I specified the right model. Factor C is indeed nested within factor A. I think you were confused by the numbering of C (1..11), and it is easier to understand when I code it as you suggested (1,2,3 within each level of A, as in mydata1 [see below]). However, it does not matter which numbering I choose for carrying the analysis, as anova(lm(resp ~ A * B + (C %in% A), mydata)) anova(lm(resp ~ A * B + (C %in% A), mydata1)) both give the same results (as at least I had expected because of the nesting). However, I found that Anova() from the car package only accepts the second version. So, Anova(lm(resp ~ A * B + (C %in% A), mydata)) does not work (giving an error) but Anova(lm(resp ~ A * B + (C %in% A), mydata1)) does. This behaviour is rather confusing, or is there anything I'm missing? You're not listening to what I told you! A term C %in% A (or A/C) is not a _specification_ that C is nested in A, it is a _directive_ to include the terms A and C:A. Now, C:A involves a term for each combination of A and C, of which many are empty if C is strictly coarser than A. This may well be what is confusing Anova(). In fact, with this (c(1:3,6:11)) coding of C, A:C is completely equivalent to C, but if you look at summary(lm()) you will see a lot of NA coefficients in the A:C case. If you use resp ~ A*B+C, then you still get a couple of missing coefficients in the C terms because of collinearity with the A terms. (Notice that this is one case where the order inside the model formula will matter; C+A*B is not the same.) Whether you'd want C as a random factor is a different matter. It is often the natural model if C is subject and A is group. Let's assume that this is the case: In an ordinary linear model, you can test whether you can remove C (or A:C) , which implies that all subjects in the same group have the same level of the response. In your case, the hypothesis is accepted, but the F statistic is around 3 (on (6, 6) DF) , which suggests that there might be some variation of subjects within groups. In a mixed-effects model, you assume that this variation exists and therefore you use the SSD for C as the denominator when testing A, which is arguably safer than pooling it with the somewhat smaller residual SSD. Thanks for your help again, Carsten R mydata A B C resp 1 1 1 1 34.12 2 1 1 2 32.45 3 1 1 3 44.55 4 1 2 1 20.88 5 1 2 2 22.32 6 1 2 3 27.71 7 2 1 6 38.20 8 2 1 7 31.62 9 2 1 8 38.71 102 2 6 18.93 112 2 7 20.57 122 2 8 31.55 133 1 9 40.81 143 1 10 42.23 153 1 11 41.26 163 2 9 28.41 173 2 10 24.07 183 2 11 21.16 R mydata1 A B C resp 1 1 1 1 34.12 2 1 1 2 32.45 3 1 1 3 44.55 4 1 2 1 20.88 5 1 2 2 22.32 6 1 2 3 27.71 7 2 1 1 38.20 8 2 1 2 31.62 9 2 1 3 38.71 102 2 1 18.93 112 2 2 20.57 122 2 3 31.55 133 1 1 40.81 143 1 2 42.23 153 1 3 41.26 163 2 1 28.41 173 2 2 24.07 183 2 3 21.16 On Tue, 2007-07-10 at 13:54 +0200, Peter Dalgaard wrote: Carsten Jaeger wrote: Hello, is it possible to obtain type III sums of squares for a nested model as in the following: lmod - lm(resp ~ A * B + (C %in% A), mydata)) I have tried library(car) Anova(lmod, type=III) but this gives me an error (and I also understand from the documentation of Anova as well as from a previous request (http://finzi.psych.upenn.edu/R/Rhelp02a/archive/64477.html) that it is not possible to specify nested models with car's Anova). anova(lmod) works, of course. My data (given below) is balanced so I expect the results to be similar for both type I and type III sums of squares. But are they *exactly* the same? The editor of the journal which I'm sending my manuscript to requests what he calls conventional type III tests and I'm not sure if can convince him to accept my type I analysis. In balanced designs, type I-IV SSD's are all identical. However, I don't think the model does what I think you think it does. Notice that nesting is used with two diferent meanings, in R it would be that the codings of C only makes sense within levels of A - e.g. if they were numbered 1:3 within each group, but with C==1 when A==1 having nothing to do with C==1 when A==2. SAS does something. er. else... What I think you want is a model where C is a random terms so that main effects of A can be tested, like in summary(aov(resp ~ A * B + Error(C), dd))
Re: [R] elementary statistics with R (rkward?)
Christopher W. Ryan wrote: As a fellow beginner, I also found Handbook of Statistical Analyses Using R, by Brian Everitt, to be a very useful book. There is an accompanying R package, HSAUR. Also Using R for Introductory Statistics, by John Verzani. There is an accompanying R package, UsingR. And for Peter Dalgaard's book there is the ISwR package: http://cran.r-project.org/src/contrib/Descriptions/ISwR.html Christopher W. Ryan, MD SUNY Upstate Medical University Clinical Campus at Binghamton 40 Arch Street, Johnson City, NY 13790 cryanatbinghamtondotedu PGP public keys available at http://home.stny.rr.com/ryancw/ If you want to build a ship, don't drum up the men to gather wood, divide the work and give orders. Instead, teach them to yearn for the vast and endless sea. [Antoine de St. Exupery] Charles Annis, P.E. wrote: Face the music and buy the book: _Introductory Statistics with R_ by Peter Dalgaard. It's perfect for what you need. It's clear and concise and will teach you statistics AND R as painlessly as such a thing can be. It's inexpensive and you can get it on Amazon.com and every other major bookseller, including the nearest university bookstore. Charles Annis, P.E. [EMAIL PROTECTED] phone: 561-352-9699 eFax: 614-455-3265 http://www.StatisticalEngineering.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Donatas G. Sent: Wednesday, July 11, 2007 9:27 AM To: r-help@stat.math.ethz.ch Subject: [R] elementary statistics with R (rkward?) Hi, I am trying to learn some basic statistics stuff but I cannot find any elementary statistics exercises using R language. Using RKward would be even better... I need that in analysing sociological data, obtained through questionnairres - findind corelations between variables, relations between different types of data, etc. Could anyone recommend simple tutorials/exercises, available on www for me to work on? I realize it would be much simple to do this introductory stuff with spss, that everyone around me is using here in Lithuania, but I'd really like to learn to do it with R instead... __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Chuck Cleland, Ph.D. NDRI, Inc. 71 West 23rd Street, 8th floor New York, NY 10010 tel: (212) 845-4495 (Tu, Th) tel: (732) 512-0171 (M, W, F) fax: (917) 438-0894 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] p-value from survreg(), library(survival)
On Wed, 2007-07-11 at 16:41 +0200, hadley wickham wrote: On 7/11/07, Marc Schwartz [EMAIL PROTECTED] wrote: Actually, in this case, looking at the code for: survival:::print.survreg would be better, as the p value is calculate there, rather than being part of the survreg object. As with many R functions, the p value is calculated in the print method for the object. I wish print methods wouldn't do that. Printing is supposed to be about displaying existing create, not creating new values. Hadley It has been occasionally confusing and I am not sure of the history behind the diverse approach. To borrow a phrase from the DoD TCSEC books[1], I don't have my S/R Rainbow Books at hand to research it. Both sets of colors are on shelves in my home office. The first time I came across this years ago, was with the p value for the F statistic in a simple linear model. it is calculated in: stats:::print.summary.lm while the individual term p values are calculated in summary.lm versus being part of the returned lm object itself. I have just become behaviorally modified to look in more than one place for such things... :-) Regards, Marc [1] http://en.wikipedia.org/wiki/Rainbow_Series __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] 3D plot and interactive PDFs
Thanks vito I was aware of movie15 but the point is how to get a U3D or VRLM file out of R :/ I don't know those two standards, nor I know the usual format for R 3D plots ... And unfortunately I am a bit in a rush so no way, right now, to do some reverse engineering about plot file format in order to convert them in VRLM :D This does not answer exactly to your question, anyway..: If you are planning to use latex, the package movie15 allows to include media files in your document (to be processed via pdflatex) vito Bruno C. wrote: With version 8 of acrobat reader, it is now possible to have 3D in PDf documents. Does it exist already an R package who manage to produce 3D plots which can be saved as interactive 3D graphs in a PDF file? Best Regards Bruno Cavestro -- Leggi GRATIS le tue mail con il telefonino i-modeTM di Wind http://i-mode.wind.it/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Vito M.R. Muggeo Dip.to Sc Statist e Matem `Vianelli' Università di Palermo viale delle Scienze, edificio 13 90128 Palermo - ITALY tel: 091 6626240 fax: 091 485726/485612 -- Leggi GRATIS le tue mail con il telefonino i-mode di Wind http://i-mode.wind.it/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] tkfocus issue
Dear All: I am stuck with this issue: I have a button on a TK window, once click it, it pops up a individual plot device: individual_plot - function() { tkconfigure(overlay.button, state=normal) options(locatorBell = FALSE) plotfuntype() trellis.focus(panel, 1, 1,highlight=FALSE) panel.identify(labels=colnames(dataplot)) } Now I have another button, originally state=disabled, but activated by : tkconfigure(overlay.button, state=normal) in previous function: The overlay function was writen to overlay another plot to the original plot. --The problem: 1.once the plot device is out, I can't go back to the Tk window, it is take hostage by the plot image some how. As a result, overlay button won't work. 2. Is there a mechanism to detect close/destroy of the plotting device in R? I want to use that even to make overlay.button state to disabled so user won't be able to click overlay just to get an error message if there is no plot already exist... Thanks Hao __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] exces return by mktcap decile for each year
here is one way of doing it using 'ave': dat - read.table(textConnection(mc yrret + 32902.233 01/01/1995 0.426 + 15793.691 01/01/1995 0.024 + 2375.868 01/01/1995 0.660 + 54586.558 01/01/1996 0.497 + 10674.900 01/01/1996 0.405 +859.656 01/01/1996 -0.033 +770.963 01/01/1995 -1.248 +423.480 01/01/1995 0.654 + 2135.504 01/01/1995 0.394 +696.599 01/01/1995 -0.482 + 5115.476 01/01/1995 0.352 +821.347 01/01/1995 0.869 + 43329.695 01/01/1995 0.495 + 7975.151 01/01/1995 0.112 +396.450 01/01/1995 0.956 +843.870 01/01/1995 0.172 + 2727.037 01/01/1995 -0.358 +114.584 01/01/1995 -1.015 + 1347.327 01/01/1995 -0.083 + 4592.049 01/01/1995 -0.251 +674.305 01/01/1995 -0.327 + 39424.887 01/01/1996 0.198 + 4447.383 01/01/1996 -0.045 + 1608.540 01/01/1996 -0.109 +217.151 01/01/1996 0.539 + 1813.320 01/01/1996 0.754 +145.170 01/01/1996 0.249 + 3176.298 01/01/1996 -0.202 + 14379.686 01/01/1996 0.013 + 3009.059 01/01/1996 -0.328 + 1781.406 01/01/1996 -0.158 + 2576.215 01/01/1996 0.514 + 1236.317 01/01/1996 0.346 + 3003.735 01/01/1996 0.151 + 1544.003 01/01/1996 0.482 + 7588.657 01/01/1996 0.306 + 1516.625 01/01/1996 0.183 + 1596.098 01/01/1996 0.674 + 2792.192 01/01/1996 0.528 + 1276.702 01/01/1996 0.010 +875.716 01/01/1996 0.189 + 4858.450 01/01/1995 0.250 + 2033.623 01/01/1995 -0.582 + 2164.125 01/01/1995 0.631), header=TRUE) # quantiles by year (need as grouping in next statement dat$qByYr - ave(dat$mc, dat$yr, FUN=function(x){ + cut(x, quantile(x, prob=seq(0, 1, .1)), include.lowest=TRUE) + }) # compute the mean for year/quantile dat$dec.mean - ave(dat$ret, dat$yr, dat$qByYr, FUN=mean) # mean adjusted return dat$mean.adjusted - dat$ret - dat$dec.mean dat mc yrret qByYr dec.mean mean.adjusted 1 32902.233 01/01/1995 0.42610 0.4605000 -0.03450 2 15793.691 01/01/1995 0.024 9 0.068 -0.04400 3 2375.868 01/01/1995 0.660 6 0.6455000 0.01450 4 54586.558 01/01/1996 0.49710 0.236 0.26100 5 10674.900 01/01/1996 0.405 9 0.3555000 0.04950 6859.656 01/01/1996 -0.033 1 0.2516667 -0.28467 7770.963 01/01/1995 -1.248 3 -0.1895000 -1.05850 8423.480 01/01/1995 0.654 1 0.198 0.45567 9 2135.504 01/01/1995 0.394 5 -0.094 0.48800 10 696.599 01/01/1995 -0.482 2 -0.4045000 -0.07750 11 5115.476 01/01/1995 0.352 8 0.301 0.05100 12 821.347 01/01/1995 0.869 3 -0.1895000 1.05850 13 43329.695 01/01/1995 0.49510 0.4605000 0.03450 14 7975.151 01/01/1995 0.112 9 0.068 0.04400 15 396.450 01/01/1995 0.956 1 0.198 0.75767 16 843.870 01/01/1995 0.172 4 0.0445000 0.12750 17 2727.037 01/01/1995 -0.358 7 -0.3045000 -0.05350 18 114.584 01/01/1995 -1.015 1 0.198 -1.21333 19 1347.327 01/01/1995 -0.083 4 0.0445000 -0.12750 20 4592.049 01/01/1995 -0.251 7 -0.3045000 0.05350 21 674.305 01/01/1995 -0.327 2 -0.4045000 0.07750 22 39424.887 01/01/1996 0.19810 0.236 -0.03800 23 4447.383 01/01/1996 -0.045 8 -0.1235000 0.07850 24 1608.540 01/01/1996 -0.109 5 0.162 -0.27133 25 217.151 01/01/1996 0.539 1 0.2516667 0.28733 26 1813.320 01/01/1996 0.754 5 0.162 0.59167 27 145.170 01/01/1996 0.249 1 0.2516667 -0.00267 28 3176.298 01/01/1996 -0.202 8 -0.1235000 -0.07850 29 14379.686 01/01/1996 0.01310 0.236 -0.22300 30 3009.059 01/01/1996 -0.328 7 -0.0885000 -0.23950 31 1781.406 01/01/1996 -0.158 5 0.162 -0.32033 32 2576.215 01/01/1996 0.514 6 0.521 -0.00700 33 1236.317 01/01/1996 0.346 2 0.2675000 0.07850 34 3003.735 01/01/1996 0.151 7 -0.0885000 0.23950 35 1544.003 01/01/1996 0.482 4 0.578 -0.09600 36 7588.657 01/01/1996 0.306 9 0.3555000 -0.04950 37 1516.625 01/01/1996 0.183 3 0.0965000 0.08650 38 1596.098 01/01/1996 0.674 4 0.578 0.09600 39 2792.192 01/01/1996 0.528 6 0.521 0.00700 40 1276.702 01/01/1996 0.010 3 0.0965000 -0.08650 41 875.716 01/01/1996 0.189 2 0.2675000 -0.07850 42 4858.450 01/01/1995 0.250 8 0.301 -0.05100 43 2033.623 01/01/1995 -0.582 5 -0.094 -0.48800 44 2164.125 01/01/1995 0.631 6 0.6455000 -0.01450 On 7/11/07, Frank Hansen [EMAIL PROTECTED] wrote: Hi Jim, Thanks for getting back on this. I did not see your email on the help list. I or you can post this solution You are right I mis-stated about mc. mc is real, it is yr that is a factor. Here is a solution, which works, but it is clunky. I thought there might be a better/more R-like less for-loop way to
Re: [R] Power calculation for the time series experiment
The built in power functions are for the fairly straight forward situations. Yours does not appear to fit into any of those. You need to think through your problem a bit more before starting to think about power. What do you mean by effect size of 1.5 (is that 1.5 standard deviations? Or raw units? What is the SD? Is the effect of 1.5 the same at each time point? Or would it change?) How do you plan on analyzing the data? Manova? Lme? What do you expect the correlation structure to be? I would suggest creating a dataset that represents the structure that you expect (includes the time points, treatment group, and any thing else). Then fill in the response with random data (rnorm to start, mvrnorm may be useful for the correlated part). Now analyze this data with the tool you plan to use to make sure that it works and gives the expected output. Now take the code you used above and create a function or set of lines such that it is easy to change things like the overall sample size, the correlation(s), the SD and/or effect size. Have the result of the function or code be the p-value of interest. Now use the replicate function to run this code/function a bunch of times, the number of times that the p-value is less than your alpha is your estimate of the power for that set of conditions. Now change some conditions (sample size, correlation, ...) and repeate. Hope this helps, -- Gregory (Greg) L. Snow Ph.D. Statistical Data Center Intermountain Healthcare [EMAIL PROTECTED] (801) 408-8111 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of A Ezhil Sent: Wednesday, July 11, 2007 8:15 AM To: r-help@stat.math.ethz.ch Subject: [R] Power calculation for the time series experiment Hi All, We are planning to run an experiment, where samples will be taken at different time points (say, 0, 4, 8, 16, 24). If I am interested in the effect size of 1.5 for a reasonably large samples (say 500), what will be the power? Is it a good idea to use F-test (one-way ANOVA) as my test statistics? How can we include correlation structure among samples in the power analysis, if I use one-way ANOVA design? I am aware of power.anova.test() in R that will help me to do power calculation for one-way ANOVA. It will be of great help if you send me some related articles or pointers to some useful resources. Thanks in advance. Kind regards, Ezhil __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] survfit for interval censored data
Sandra, As far as I am aware, you will have to use a parametric model (survreg) if your survival times are interval-censored. Regards, -Cody Cody Hamilton Edwards Lifesciences -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Sandra Ellis Sent: Wednesday, July 11, 2007 7:11 AM To: r-help@stat.math.ethz.ch Subject: [R] survfit for interval censored data pHello,/ppI am a new R-user and would like to use survfit for interval censored data. Whenever I try I get an error message that states I can only use survfit for right censored or counting process data. I was wondering if anyone knows if there is an additional package available that can calculate KM curves for interval censored data, or another program with this capability?/ppThank you./ppS. Ellis (student)/p __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Drawing rectangles in multiple panels
Hi folks, I'm having some trouble understanding the intricacies of panel functions. I wish to create three side-by-side graphs, each with different data-- so far, so good: I rbind() the data, add a column of subscripts as a conditioning variable, load up the lattice package, specify either a c(3,1) 'layout' or work through 'allow.multiple' and 'outer' and I'm good to go. But now I wish to add three rectangles to each plot, which will be in different places on each panel, and I'm terribly stuck. I can guess this requires defining a panel function on the fly, but none of my attempts are working. Suggestions? Thanks much, - Jonathan __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Drawing rectangles in multiple panels
On 7/11/07, Jonathan Williams [EMAIL PROTECTED] wrote: Hi folks, I'm having some trouble understanding the intricacies of panel functions. I wish to create three side-by-side graphs, each with different data-- so far, so good: I rbind() the data, add a column of subscripts as a conditioning variable, load up the lattice package, specify either a c(3,1) 'layout' or work through 'allow.multiple' and 'outer' and I'm good to go. But now I wish to add three rectangles to each plot, which will be in different places on each panel, and I'm terribly stuck. I can guess this requires defining a panel function on the fly, but none of my attempts are working. Suggestions? You haven't told us what determines the rectangles (only that they are different in each panel). If they are completely driven by panel data, here's an example: panel.qrect - function(x, y, ...) { xq - quantile(x, c(0.1, 0.9)) yq - quantile(y, c(0.1, 0.9)) panel.rect(xq[1], yq[1], xq[2], yq[2], col = grey86, border = NA) panel.xyplot(x, y, ...) } xyplot(Sepal.Length ~ Sepal.Width | Species, iris, panel = panel.qrect) If the rectangles are somehow determined externally, you probably want to use one of the accessor functions described in help(panel.number). There are good and bad (i.e. less robust) ways to use these, but we need to know your use case before recommending one. -Deepayan __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] error using lp function in linux
Perhaps I should clarify that when I use sample code from the lp help function, I can paste it and run it in my windows-based R with no problems. My question is why the same code won't work on a linux system. Thanks, Byran Smucker On Wed, 11 Jul 2007 12:00:09 +0200 r-help@stat.math.ethz.ch wrote: Message: 56 Date: Tue, 10 Jul 2007 19:07:10 -0400 From: quot;Byran Smuckerquot; [EMAIL PROTECTED] Subject: [R] error using lp function in linux To: r-help@stat.math.ethz.ch Message-ID: [EMAIL PROTECTED] Content-Type: text/plain Hello all, I would like to use the lp function (lpSolve package) on a linux system. Using sample code from the lp function help, I can solve the linear program with no errors or problems. However, when I copy the exact same code to R in linux (after loading the lpSolve package), I get the following error: *** caught segfault *** address 0x3, cause 'memory not mapped' Traceback: 1: .C(quot;lpslinkquot;, direction = as.integer(direction), x.count = as.integer(x.count), objective = as.double(objective), const.count = as.integer(const.count), constraints = as.double(constraints), int.count = as.integer(int.count), int.vec = as.integer(int.vec), objval = as.double(objval), solution = as.double(solution), presolve = as.integer(presolve), compute.sens = as.integer(compute.sens), sens.coef.from = as.double(sens.coef.from), sens.coef.to = as.double(sens.coef.to), duals = as.double(duals), duals.from = as.double(duals.from), duals.to = as.double(duals.to), status = as.integer(status), PACKAGE = quot;lpSolvequot;) 2: lp(quot;maxquot;, f.obj, f.con, f.dir, f.rhs) Possible actions: 1: abort (with core dump, if enabled) 2: normal R exit 3: exit R without saving workspace 4: exit R saving workspace Does this function not work with linux or is there some other explanation?Thanks, Byran Smucker Graduate Student, Dept. of Statistics Penn State University 301 Thomas Building [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Drawing rectangles in multiple panels
Deepayan, Thanks for the clarification. The rectangles are completely external to the panel data, and correspond to 90% confidence intervals built from training data, to be overlaid on these graphs of the test data. - Jonathan At 10:04 AM 7/11/2007, you wrote: On 7/11/07, Jonathan Williams [EMAIL PROTECTED] wrote: Hi folks, I'm having some trouble understanding the intricacies of panel functions. I wish to create three side-by-side graphs, each with different data-- so far, so good: I rbind() the data, add a column of subscripts as a conditioning variable, load up the lattice package, specify either a c(3,1) 'layout' or work through 'allow.multiple' and 'outer' and I'm good to go. But now I wish to add three rectangles to each plot, which will be in different places on each panel, and I'm terribly stuck. I can guess this requires defining a panel function on the fly, but none of my attempts are working. Suggestions? You haven't told us what determines the rectangles (only that they are different in each panel). If they are completely driven by panel data, here's an example: panel.qrect - function(x, y, ...) { xq - quantile(x, c(0.1, 0.9)) yq - quantile(y, c(0.1, 0.9)) panel.rect(xq[1], yq[1], xq[2], yq[2], col = grey86, border = NA) panel.xyplot(x, y, ...) } xyplot(Sepal.Length ~ Sepal.Width | Species, iris, panel = panel.qrect) If the rectangles are somehow determined externally, you probably want to use one of the accessor functions described in help(panel.number). There are good and bad (i.e. less robust) ways to use these, but we need to know your use case before recommending one. -Deepayan __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Drawing rectangles in multiple panels
Deepayan et. al.: A question/comment: I have usually found that the subscripts argument is what I need when passing *external* information into the panel function, for example, when I wish to add results from a fit done external to the trellis call. Fits[subscripts] gives me the fits (or whatever) I want to plot for each panel. It is not clear to me how the panel layout information from panel.number(), etc. would be helpful here instead. Am I correct? -- or is there a smarter way to do this that I've missed? Cheers, Bert Bert Gunter Genentech Nonclinical Statistics -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED] Sent: Wednesday, July 11, 2007 10:04 AM To: Jonathan Williams Cc: r-help@stat.math.ethz.ch Subject: Re: [R] Drawing rectangles in multiple panels On 7/11/07, Jonathan Williams [EMAIL PROTECTED] wrote: Hi folks, I'm having some trouble understanding the intricacies of panel functions. I wish to create three side-by-side graphs, each with different data-- so far, so good: I rbind() the data, add a column of subscripts as a conditioning variable, load up the lattice package, specify either a c(3,1) 'layout' or work through 'allow.multiple' and 'outer' and I'm good to go. But now I wish to add three rectangles to each plot, which will be in different places on each panel, and I'm terribly stuck. I can guess this requires defining a panel function on the fly, but none of my attempts are working. Suggestions? You haven't told us what determines the rectangles (only that they are different in each panel). If they are completely driven by panel data, here's an example: panel.qrect - function(x, y, ...) { xq - quantile(x, c(0.1, 0.9)) yq - quantile(y, c(0.1, 0.9)) panel.rect(xq[1], yq[1], xq[2], yq[2], col = grey86, border = NA) panel.xyplot(x, y, ...) } xyplot(Sepal.Length ~ Sepal.Width | Species, iris, panel = panel.qrect) If the rectangles are somehow determined externally, you probably want to use one of the accessor functions described in help(panel.number). There are good and bad (i.e. less robust) ways to use these, but we need to know your use case before recommending one. -Deepayan __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] How to load permtest package
Hi sir Sir I am not understanding how to load permtest on R console for windows? Regards -- AMINA SHAHZADI Department of Statistics GC University Lahore, Pakistan. Email: [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] aov() question
Hi all, So I think I have seen some similar questions to mine when I searched the archives, but have not seen any concrete answers and was wondering if any one could help. I have been trying to use R's aov() function to analyze my data. I have a 3 x 4 x 2 repeated measures design. All of the IVs are within subjects.I do also have missing values (unequal N), as I have to remove any incorrect trials for each subject. Here is the code I entered and the error message: a-aov(log(rt)~(tran*block*half) + Error (sid/ (tran*block*half)), data=mydata2) Warning message: Error() model is singular in: aov(log(rt) ~ (tran * block * half) + Error(sid/(tran * block * I then do summary(a) and am able to get an output, but I am not sure whether or not I can trust that output since I got the error message. Any body have any thoughts/solutions for this? Also, are there any benefits of you aov() vs. use some of the linear model functions or vice versa? Thanks for any help you can offer!! ~Leigh Alexander __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Drawing rectangles in multiple panels
On 7/11/07, Bert Gunter [EMAIL PROTECTED] wrote: Deepayan et. al.: A question/comment: I have usually found that the subscripts argument is what I need when passing *external* information into the panel function, for example, when I wish to add results from a fit done external to the trellis call. Fits[subscripts] gives me the fits (or whatever) I want to plot for each panel. It is not clear to me how the panel layout information from panel.number(), etc. would be helpful here instead. Am I correct? -- or is there a smarter way to do this that I've missed? subscripts is absolutely the right thing to use if your auxiliary information is in the form of vectors than have the same length as the rest of your data. Examples would include a color for every point in a xyplot or confidence bounds in a dotplot. However, sometimes your external information might be a summary; say the parameters defining a fitted curve for every combination of your conditioning variables (and the underlying model might have shared information across combinations, so you wouldn't be able to compute them from the panel data alone). In that case, which.packet(), which gives you the levels of the conditioning variables defining the current panel, may be helpful. Of course, there are workarounds using subscripts, or in this example, adding rows to your data frame containing the fitted values. The accessors are a convenience that sometimes make life simpler. -Deepayan __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Drawing rectangles in multiple panels
On 7/11/07, Jonathan Williams [EMAIL PROTECTED] wrote: Deepayan, Thanks for the clarification. The rectangles are completely external to the panel data, and correspond to 90% confidence intervals built from training data, to be overlaid on these graphs of the test data. Right. So if you have that information in a single object (say a list), you can specify that as an argument to xyplot (or whatever), and capture that in the panel function to then index it. E.g. rectInfo - list(matrix(runif(4), 2, 2), matrix(runif(4), 2, 2), matrix(runif(4), 2, 2)) panel.qrect - function(x, y, ..., rect.info) { ri - rect.info[[packet.number()]] ## if you have more than one conditioning variable, this might be ## something like ## ri - do.call([, list(rect.info, as.list(which.packet([[1]] panel.rect(ri[1, 1], ri[1, 2], ri[2, 1], ri[2, 2], col = grey86, border = NA) panel.xyplot(x, y, ...) } xyplot(runif(30) ~ runif(30) | gl(3, 10), rect.info = rectInfo, panel = panel.qrect) -Deepayan __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] system: Linux vs. Windows differences
[I tried to send this messages two days ago, but I guess I mistyped the To: address...] Why system is different in Linux and Windows? Both in R 2.4.1, but in Windows there is an option: system(something, wait = FALSE) while on Linux (Fedora Core 4), there is no such option? Alberto Monteiro __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Drawing rectangles in multiple panels
A question/comment: I have usually found that the subscripts argument is what I need when passing *external* information into the panel function, for example, when I wish to add results from a fit done external to the trellis call. Fits[subscripts] gives me the fits (or whatever) I want to plot for each panel. It is not clear to me how the panel layout information from panel.number(), etc. would be helpful here instead. Am I correct? -- or is there a smarter way to do this that I've missed? This is one of things that I think ggplot does better - it's much easier to plot multiple data sources. I don't have many examples of this yet, but the final example on http://had.co.nz/ggplot2/geom_abline.html illustrates the basic idea. For the original poster ggplot2 isn't that much more convenient, because there isn't a built in rectangle geom (although it would be trivial to add one). You could use the more general polygon geom, http://had.co.nz/ggplot2/geom_polygon.html, however it currently doesn't have a lot of documentation. Hadley __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] system: Linux vs. Windows differences
On Wed, 2007-07-11 at 16:16 -0200, Alberto Monteiro wrote: [I tried to send this messages two days ago, but I guess I mistyped the To: address...] Why system is different in Linux and Windows? Both in R 2.4.1, but in Windows there is an option: system(something, wait = FALSE) while on Linux (Fedora Core 4), there is no such option? Alberto Monteiro From: https://svn.r-project.org/R/trunk/NEWS for changes in R 2.5.0: o system() now takes the same set of arguments on all platforms, with those which are not applicable being ignored with a warning. Unix-alikes gain 'input' and 'wait', and Windows gains 'ignore.stderr'. Time to upgrade both your R and your FC installation. R is at 2.5.1. FC4 has been EOL (End of Life) for some time and FC5 hit EOL last month. HTH, Marc Schwartz __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Stepwise GLM selection by LRT?
Dear List, having searched the help and archives, I have the impression that there is no automatic model selection procedure implemented in R that includes/excludes predictors in logistic regression models based on LRT P-values. Is that true, or is someone aware of an appropriate function somewhere in a custom package? Even if automatic model selection and LRT might not be the most appropriate methods, I actually would like to use these in order to simulate someone else's modeling approach... Many thanks for all comments- Lutz - Lutz Ph. Breitling German Cancer Research Center Heidelberg/Germany __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Stepwise GLM selection by LRT?
Check out the stepAIC function in MASS package. This is a nice tool, where you can actually implement any penalty even though the function's name has AIC in it because it is the default. Although this doesn't do an LRT test based variable selection, you can sort of approximate it by using a penalty of k = qchisq(1-p, df=1), where p is the p-value for variable selection. This penalty means that a variable enters/exits an existing model, when the absolute value of change in log-likelihood is greater than qchisq(1-p, df=1). For p = 0.1, k = 2.71, and for p=0.05, k = 3.84. Is this whhant you'd like to do? Ravi. --- Ravi Varadhan, Ph.D. Assistant Professor, The Center on Aging and Health Division of Geriatric Medicine and Gerontology Johns Hopkins University Ph: (410) 502-2619 Fax: (410) 614-9625 Email: [EMAIL PROTECTED] Webpage: http://www.jhsph.edu/agingandhealth/People/Faculty/Varadhan.html -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Lutz Ph. Breitling Sent: Wednesday, July 11, 2007 3:06 PM To: r-help@stat.math.ethz.ch Subject: [R] Stepwise GLM selection by LRT? Dear List, having searched the help and archives, I have the impression that there is no automatic model selection procedure implemented in R that includes/excludes predictors in logistic regression models based on LRT P-values. Is that true, or is someone aware of an appropriate function somewhere in a custom package? Even if automatic model selection and LRT might not be the most appropriate methods, I actually would like to use these in order to simulate someone else's modeling approach... Many thanks for all comments- Lutz - Lutz Ph. Breitling German Cancer Research Center Heidelberg/Germany __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] How to get weekly Co-Variance
Hi: I am trying to migrate from Systat to R but I am facing my first challenge. While I easily can get weekly co-variance for my data, I can't seem to acomplish this with R ( I can't figure out how is done) If interested in looking a sample of my data, please check the Excel attachment. In Systat from Week 28 I get a covariance of 1055 fish. Thanks in advance Felipe D. Carrillo Fishery Biologist US Fish Wildlife Service Red Bluff, California 96080 - __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] make error R-5.1 on sun solaris
I hope this is enough information to determine the problem. Thanks in advance for any help. Configure goes ok (I think) ./configure --prefix=$HOME --without-iconv R is now configured for sparc-sun-solaris2.9 Source directory: . Installation directory:/home/dpowers C compiler:gcc -g -O2 Fortran 77 compiler: f95 -g C++ compiler: g++ -g -O2 Fortran 90/95 compiler:f95 -g Obj-C compiler: -g -O2 Interfaces supported: X11 External libraries:readline Additional capabilities: NLS Options enabled: shared BLAS, R profiling, Java Recommended packages: yes Make ends after the gcc.. make . . . gcc -I. -I../../src/include -I../../src/include -I/usr/openwin/include -I/usr/local/include -DHAVE_CONFIG_H -g -O2 -c system.c -o system.o system.c: In function `Rf_initialize_R': system.c:144: parse error before `char' system.c:216: `localedir' undeclared (first use in this function) system.c:216: (Each undeclared identifier is reported only once system.c:216: for each function it appears in.) *** Error code 1 make: Fatal error: Command failed for target `system.o' Current working directory /home/dpowers/R-2.5.1/src/unix *** Error code 1 make: Fatal error: Command failed for target `R' Current working directory /home/dpowers/R-2.5.1/src/unix *** Error code 1 make: Fatal error: Command failed for target `R' Current working directory /home/dpowers/R-2.5.1/src *** Error code 1 make: Fatal error: Command failed for target `R' I have tried setting localedir directly in configure options, but get the same error. Any ideas? Thanks, Dan =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Daniel A. Powers, Ph.D. Department of Sociology University of Texas at Austin 1 University Station A1700 Austin, TX 78712-0118 phone: 512-232-6335 fax: 512-471-1748 [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] system: Linux vs. Windows differences
On 11-Jul-07 18:28:19, Marc Schwartz wrote: On Wed, 2007-07-11 at 16:16 -0200, Alberto Monteiro wrote: [I tried to send this messages two days ago, but I guess I mistyped the To: address...] Why system is different in Linux and Windows? Both in R 2.4.1, but in Windows there is an option: system(something, wait = FALSE) while on Linux (Fedora Core 4), there is no such option? Alberto Monteiro From: https://svn.r-project.org/R/trunk/NEWS for changes in R 2.5.0: o system() now takes the same set of arguments on all platforms, with those which are not applicable being ignored with a warning. Unix-alikes gain 'input' and 'wait', and Windows gains 'ignore.stderr'. Time to upgrade both your R and your FC installation. R is at 2.5.1. FC4 has been EOL (End of Life) for some time and FC5 hit EOL last month. HTH, Marc Schwartz End of Life is a Nomenklatura categorisation. While we loyal Members welcome and applaud President-2.5.1, we do not forget old Comrades now air-brushed from the photographs who, now faceless, sturdily labour 24 hours a day in the fields and saltmines, and even in the dark attics of those who conceal and protect them still. There can be life in old dogs, and even strong teeth in some. Ted. [emailing from SuSE-5.2 (1997), logged in from Red Hat 9 (2003)] E-Mail: (Ted Harding) [EMAIL PROTECTED] Fax-to-email: +44 (0)870 094 0861 Date: 11-Jul-07 Time: 22:32:13 -- XFMail -- __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] make error R-5.1 on sun solaris
Dan Powers wrote: I hope this is enough information to determine the problem. Thanks in advance for any help. Configure goes ok (I think) ./configure --prefix=$HOME --without-iconv R is now configured for sparc-sun-solaris2.9 Source directory: . Installation directory:/home/dpowers C compiler:gcc -g -O2 Fortran 77 compiler: f95 -g C++ compiler: g++ -g -O2 Fortran 90/95 compiler:f95 -g Obj-C compiler: -g -O2 Interfaces supported: X11 External libraries:readline Additional capabilities: NLS Options enabled: shared BLAS, R profiling, Java Recommended packages: yes Make ends after the gcc.. make . . . gcc -I. -I../../src/include -I../../src/include -I/usr/openwin/include -I/usr/local/include -DHAVE_CONFIG_H -g -O2 -c system.c -o system.o system.c: In function `Rf_initialize_R': system.c:144: parse error before `char' system.c:216: `localedir' undeclared (first use in this function) system.c:216: (Each undeclared identifier is reported only once system.c:216: for each function it appears in.) *** Error code 1 make: Fatal error: Command failed for target `system.o' Current working directory /home/dpowers/R-2.5.1/src/unix *** Error code 1 make: Fatal error: Command failed for target `R' Current working directory /home/dpowers/R-2.5.1/src/unix *** Error code 1 make: Fatal error: Command failed for target `R' Current working directory /home/dpowers/R-2.5.1/src *** Error code 1 make: Fatal error: Command failed for target `R' I have tried setting localedir directly in configure options, but get the same error. Any ideas? Hmm, which version of gcc is this? The problem seems to be around line 144 which reads 140 Rstart Rp = rstart; 141 cmdlines[0] = '\0'; 142 143 #ifdef ENABLE_NLS 144 char localedir[PATH_MAX+20]; 145 #endif 146 147 #if defined(HAVE_SYS_RESOURCE_H) defined(HAVE_GETRLIMIT) 148 { 149 struct rlimit rlim; I seem to remember that it used to be non-kosher to mix declarations and ordinary code like that, but the current compiler doesn't seem to care (I do have #define ENABLE_NLS 1 in Rconfig.h, as I assume you do too). Could you perhaps try moving line 141 down below #endif? Thanks, Dan =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Daniel A. Powers, Ph.D. Department of Sociology University of Texas at Austin 1 University Station A1700 Austin, TX 78712-0118 phone: 512-232-6335 fax: 512-471-1748 [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] system: Linux vs. Windows differences
On Wed, 2007-07-11 at 22:32 +0100, [EMAIL PROTECTED] wrote: On 11-Jul-07 18:28:19, Marc Schwartz wrote: On Wed, 2007-07-11 at 16:16 -0200, Alberto Monteiro wrote: [I tried to send this messages two days ago, but I guess I mistyped the To: address...] Why system is different in Linux and Windows? Both in R 2.4.1, but in Windows there is an option: system(something, wait = FALSE) while on Linux (Fedora Core 4), there is no such option? Alberto Monteiro From: https://svn.r-project.org/R/trunk/NEWS for changes in R 2.5.0: o system() now takes the same set of arguments on all platforms, with those which are not applicable being ignored with a warning. Unix-alikes gain 'input' and 'wait', and Windows gains 'ignore.stderr'. Time to upgrade both your R and your FC installation. R is at 2.5.1. FC4 has been EOL (End of Life) for some time and FC5 hit EOL last month. HTH, Marc Schwartz End of Life is a Nomenklatura categorisation. While we loyal Members welcome and applaud President-2.5.1, we do not forget old Comrades now air-brushed from the photographs who, now faceless, sturdily labour 24 hours a day in the fields and saltmines, and even in the dark attics of those who conceal and protect them still. There can be life in old dogs, and even strong teeth in some. Ted. [emailing from SuSE-5.2 (1997), logged in from Red Hat 9 (2003)] Wow... :-) I am envisioning the above as the Prologue for a book...I am feeling suddenly melancholy... I whole heartily nominate this as a fortune! Regards, Marc __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Drawing rectangles in multiple panels
On 7/11/07, hadley wickham [EMAIL PROTECTED] wrote: A question/comment: I have usually found that the subscripts argument is what I need when passing *external* information into the panel function, for example, when I wish to add results from a fit done external to the trellis call. Fits[subscripts] gives me the fits (or whatever) I want to plot for each panel. It is not clear to me how the panel layout information from panel.number(), etc. would be helpful here instead. Am I correct? -- or is there a smarter way to do this that I've missed? This is one of things that I think ggplot does better - it's much easier to plot multiple data sources. I don't have many examples of this yet, but the final example on http://had.co.nz/ggplot2/geom_abline.html illustrates the basic idea. That's probably true. The Trellis approach is to define a plot by data source + type of plot, whereas the ggplot approach (if I understand correctly) is to create a specification for the display (incrementally?) and then render it. Since the specification can be very general, the approach is very flexible. The downside is that you need to learn the language. On a philosophical note, I think the apparent limitations of Trellis in some (not all) cases is just due to the artificial importance given to data frames as the one true container for data. Now that we have proper multiple dispatch in S4, we can write methods that behave like traditional Trellis calls but work with more complex data structures. We have tried this in one bioconductor package (flowViz) with encouraging results. -Deepayan __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] type III ANOVA for a nested linear model
Mark Difford wrote: Indeed! And, apropos of the expression, to be Ripleyed (and so be condemned to eating cookies for a long, long time), what about being Billasted? BestR, Mark. Simon Blomberg-4 wrote: I second the nomination! Simon. On Tue, 2007-07-10 at 10:02 -0600, Greg Snow wrote: I nominate the following 2 pieces from Bill's reply for fortunes (probably 2 separate fortunes): All this becomes even more glaring if you take the unusal step of plotting the data. and What sort of editor would overlook this clear and demonstrable message leaping out from the data in favour of some arcane argument about types of sums of squares? Several answers come to mind: A power freak, a SAS afficianado, an idiot. More seriously on this topic is the need to educate editors. Few editors in the biological field (real biology, biomed etc) appear to have, or if they have, to exercise, any sort of near current judgment on statistical methods, techniques or interpretations. Referees often have no knowledge of analysis and editors blindly back their referee ... (my own pet gripe here is being asked for replicates when showing the existence of a phenomenon ... The assertion 'all swans are black' is refuted by the observation of a white swan ... Referee: how many replicates did the researcher have? There appears to be a single sample here; where is the confidence interval on the proportion of white swans?). A little learning is a dangerous thing ... (Pope) ... and most biological editors may have a compulsory undergraduate subject in their distant background, from which they remember 'Yates' correction' and 'no cell with fewer than 5 observations' (sic), Richard Rowe -- Dr Richard Rowe Zoology Tropical Ecology School of Tropical Biology James Cook University Townsville 4811 AUSTRALIA ph +61 7 47 81 4851 fax +61 7 47 25 1570 JCU has CRICOS Provider Code 00117J __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] rgdal memory error for a small map.
Dear all, I am trying to read a - almost for me - small map on R using rgdal package. This image have dimension of 1701x1401 and are stored on native ArcGis GRID format. When I see the files sizes, it is less then one megabyte. But unfortunately when I try read using readGDAL function it return a memory.size() error. I don´t know why it occours, if it is a small map. Below you can see GDALinfo(), readGDAL() = with the error, sessionInfo() and memory.size() outputs. Any idea? Kind regards, miltinho Brazil GDALinfo('al001_frag') Closing GDAL dataset handle 0x0134d310... destroyed ... done. rows1701 columns 1401 bands 1 ll.x0.5 ll.y1701.5 res.x 1 res.y 1 oblique.x 0 oblique.y 0 driver AIG projection NA fileal001_frag tst-readGDAL(al001_frag) al001_frag has GDAL driver AIG and has 1701 rows and 1401 columns Closing GDAL dataset handle 0x020d7c68... destroyed ... done. Warning messages: 1: Reached total allocation of 479Mb: see help(memory.size) 2: Reached total allocation of 479Mb: see help(memory.size) * sessionInfo() R version 2.5.0 (2007-04-23) i386-pc-mingw32 locale: LC_COLLATE=English_Jamaica.1252;LC_CTYPE=English_Jamaica.1252;LC_MONETARY=English_Jamaica.1252;LC_NUMERIC=C;LC_TIME=English_Jamaica.1252 attached base packages: [1] stats graphics grDevices utils datasets methods [7] base other attached packages: rgdal sp 0.5-13 0.9-14 memory.size() [1] 218914992 http://yahoo.com.br/oqueeuganhocomisso [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Drawing rectangles in multiple panels
Not that Trellis/lattice was entirely easy to learn at first. :) I've been playing around with ggplot2 and there is a plot()-like wrapper for building a quick plot [incidentally, called qplot()], but otherwise it's my understanding that you superpose elements (incrementally) to build up to the graph you want. Here is the same plot in ggplot2: rectInfo - list(matrix(runif(4), 2, 2), matrix(runif(4), 2, 2), matrix(runif(4), 2, 2)) library(ggplot2) ggopt(grid.fill = white) # just my preference ## original plot of points p - qplot(x,y,data=data.frame(x=runif(30),y=runif(30),f=gl(3,30)),facets=f~.) # print(p) ## external data (rectangles) - in coordinates for geom_polygon x - do.call(rbind, mapply(function(.r,.f) data.frame(x=.r[c(1,1,2,2),1],y=.r[c(1,2,2,1),2],f=.f), .r=rectInfo,.f=seq(along=rectInfo),SIMPLIFY=FALSE)) ## add rectangle to original plot of points p+layer(geom=polygon,data=x,mapping=aes(x=x,y=y),facets=f~.) # will print the graphics on my windows() device Though lattice does seem to emphasize the 'chart type' approach to graphing, in a way I see that it provides a similar flexibility - just that the specifications for each element are contained in functions and objects that are ultimately invoked by a high-level/higher-order function, instead of being combined in the linear fashion of ggplot2. ST --- Deepayan Sarkar [EMAIL PROTECTED] wrote: On 7/11/07, hadley wickham [EMAIL PROTECTED] wrote: A question/comment: I have usually found that the subscripts argument is what I need when passing *external* information into the panel function, for example, when I wish to add results from a fit done external to the trellis call. Fits[subscripts] gives me the fits (or whatever) I want to plot for each panel. It is not clear to me how the panel layout information from panel.number(), etc. would be helpful here instead. Am I correct? -- or is there a smarter way to do this that I've missed? This is one of things that I think ggplot does better - it's much easier to plot multiple data sources. I don't have many examples of this yet, but the final example on http://had.co.nz/ggplot2/geom_abline.html illustrates the basic idea. That's probably true. The Trellis approach is to define a plot by data source + type of plot, whereas the ggplot approach (if I understand correctly) is to create a specification for the display (incrementally?) and then render it. Since the specification can be very general, the approach is very flexible. The downside is that you need to learn the language. On a philosophical note, I think the apparent limitations of Trellis in some (not all) cases is just due to the artificial importance given to data frames as the one true container for data. Now that we have proper multiple dispatch in S4, we can write methods that behave like traditional Trellis calls but work with more complex data structures. We have tried this in one bioconductor package (flowViz) with encouraging results. -Deepayan __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Drawing rectangles in multiple panels
In the Trellis approach, another way (I like) to deal with multiple pieces of external data sources is to 'attach' them to panel functions through lexical closures. For instance... rectInfo - list(matrix(runif(4), 2, 2), matrix(runif(4), 2, 2), matrix(runif(4), 2, 2)) panel.qrect - function(rect.info) { function(x, y, ...) { ri - rect.info[[packet.number()]] panel.rect(ri[1, 1], ri[1, 2], ri[2, 1], ri[2, 2], col = grey86, border = NA) panel.xyplot(x, y, ...) } } xyplot(runif(30) ~ runif(30) | gl(3, 10), panel = panel.qrect(rectInfo)) ...which may or may not be more convenient than passing rectInfo (and perhaps other objects if desired) explicitly as an argument to xyplot(). --- Deepayan Sarkar [EMAIL PROTECTED] wrote: On 7/11/07, hadley wickham [EMAIL PROTECTED] wrote: A question/comment: I have usually found that the subscripts argument is what I need when passing *external* information into the panel function, for example, when I wish to add results from a fit done external to the trellis call. Fits[subscripts] gives me the fits (or whatever) I want to plot for each panel. It is not clear to me how the panel layout information from panel.number(), etc. would be helpful here instead. Am I correct? -- or is there a smarter way to do this that I've missed? This is one of things that I think ggplot does better - it's much easier to plot multiple data sources. I don't have many examples of this yet, but the final example on http://had.co.nz/ggplot2/geom_abline.html illustrates the basic idea. That's probably true. The Trellis approach is to define a plot by data source + type of plot, whereas the ggplot approach (if I understand correctly) is to create a specification for the display (incrementally?) and then render it. Since the specification can be very general, the approach is very flexible. The downside is that you need to learn the language. On a philosophical note, I think the apparent limitations of Trellis in some (not all) cases is just due to the artificial importance given to data frames as the one true container for data. Now that we have proper multiple dispatch in S4, we can write methods that behave like traditional Trellis calls but work with more complex data structures. We have tried this in one bioconductor package (flowViz) with encouraging results. -Deepayan __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Drawing rectangles in multiple panels
On 7/11/07, Stephen Tucker [EMAIL PROTECTED] wrote: In the Trellis approach, another way (I like) to deal with multiple pieces of external data sources is to 'attach' them to panel functions through lexical closures. For instance... rectInfo - list(matrix(runif(4), 2, 2), matrix(runif(4), 2, 2), matrix(runif(4), 2, 2)) panel.qrect - function(rect.info) { function(x, y, ...) { ri - rect.info[[packet.number()]] panel.rect(ri[1, 1], ri[1, 2], ri[2, 1], ri[2, 2], col = grey86, border = NA) panel.xyplot(x, y, ...) } } xyplot(runif(30) ~ runif(30) | gl(3, 10), panel = panel.qrect(rectInfo)) ...which may or may not be more convenient than passing rectInfo (and perhaps other objects if desired) explicitly as an argument to xyplot(). That's an interesting approach. I think the important thing is to make sure that the data required to reproduce the plot is available as part of the trellis object (e.g. if you save it and load it in another session (not that anyone actually ever does that, but it's the principle of the thing)). This happens transparently if you supply the external data as arguments to xyplot(). It happens in your example too, but the data is hidden inside the environment of the panel function. The two approaches might differ in terms of memory use, but I'm not sure. -Deepayan __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Drawing rectangles in multiple panels
Your approach of using closures is cleaner than that given below but just for comparison in: http://tolstoy.newcastle.edu.au/R/devel/06/03/4476.html there is a createWrapper function which creates a new function based on the function passed as its first argument by using the components of the list passed as its second argument to overwrite its formal arguments. For example, createWrapper - function(FUN, Params) { as.function(c(replace(formals(FUN), names(Params), Params), body(FUN))) } library(lattice) rectInfo - list(matrix(runif(4), 2, 2), matrix(runif(4), 2, 2), matrix(runif(4), 2, 2)) panel.qrect - function(x, y, ..., rect.info) { ri - rect.info[[packet.number()]] panel.rect(ri[1, 1], ri[1, 2], ri[2, 1], ri[2, 2], col = grey86, border = NA) panel.xyplot(x, y, ...) } xyplot(runif(30) ~ runif(30) | gl(3, 10), panel = createWrapper(panel.qrect, list(rect.info = rectInfo))) The createWrapper approach does have an advantage in the situation where the function analogous to panel.qrect is existing since using scoping then involves manipulation of environments in the closure approach. On 7/11/07, Stephen Tucker [EMAIL PROTECTED] wrote: In the Trellis approach, another way (I like) to deal with multiple pieces of external data sources is to 'attach' them to panel functions through lexical closures. For instance... rectInfo - list(matrix(runif(4), 2, 2), matrix(runif(4), 2, 2), matrix(runif(4), 2, 2)) panel.qrect - function(rect.info) { function(x, y, ...) { ri - rect.info[[packet.number()]] panel.rect(ri[1, 1], ri[1, 2], ri[2, 1], ri[2, 2], col = grey86, border = NA) panel.xyplot(x, y, ...) } } xyplot(runif(30) ~ runif(30) | gl(3, 10), panel = panel.qrect(rectInfo)) ...which may or may not be more convenient than passing rectInfo (and perhaps other objects if desired) explicitly as an argument to xyplot(). --- Deepayan Sarkar [EMAIL PROTECTED] wrote: On 7/11/07, hadley wickham [EMAIL PROTECTED] wrote: A question/comment: I have usually found that the subscripts argument is what I need when passing *external* information into the panel function, for example, when I wish to add results from a fit done external to the trellis call. Fits[subscripts] gives me the fits (or whatever) I want to plot for each panel. It is not clear to me how the panel layout information from panel.number(), etc. would be helpful here instead. Am I correct? -- or is there a smarter way to do this that I've missed? This is one of things that I think ggplot does better - it's much easier to plot multiple data sources. I don't have many examples of this yet, but the final example on http://had.co.nz/ggplot2/geom_abline.html illustrates the basic idea. That's probably true. The Trellis approach is to define a plot by data source + type of plot, whereas the ggplot approach (if I understand correctly) is to create a specification for the display (incrementally?) and then render it. Since the specification can be very general, the approach is very flexible. The downside is that you need to learn the language. On a philosophical note, I think the apparent limitations of Trellis in some (not all) cases is just due to the artificial importance given to data frames as the one true container for data. Now that we have proper multiple dispatch in S4, we can write methods that behave like traditional Trellis calls but work with more complex data structures. We have tried this in one bioconductor package (flowViz) with encouraging results. -Deepayan __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Drawing rectangles in multiple panels
Regarding this, I meant to imply that lattice was similarly flexible in the sense of handing multiple data sets [IMHO], in regards to other aspects of the 'grammar of graphics' I have no qualifications to justify comment. But the idea and intuitiveness of graph construction in ggplot2 is very appealing - in an hour I picked up enough to do quite a bit, just by going through examples in the author's book http://had.co.nz/ggplot2/. Will be interesting to see how this package will be received by the community. Stephen --- Stephen Tucker [EMAIL PROTECTED] wrote: Not that Trellis/lattice was entirely easy to learn at first. :) I've been playing around with ggplot2 and there is a plot()-like wrapper for building a quick plot [incidentally, called qplot()], but otherwise it's my understanding that you superpose elements (incrementally) to build up to the graph you want. Here is the same plot in ggplot2: rectInfo - list(matrix(runif(4), 2, 2), matrix(runif(4), 2, 2), matrix(runif(4), 2, 2)) library(ggplot2) ggopt(grid.fill = white) # just my preference ## original plot of points p - qplot(x,y,data=data.frame(x=runif(30),y=runif(30),f=gl(3,30)),facets=f~.) # print(p) ## external data (rectangles) - in coordinates for geom_polygon x - do.call(rbind, mapply(function(.r,.f) data.frame(x=.r[c(1,1,2,2),1],y=.r[c(1,2,2,1),2],f=.f), .r=rectInfo,.f=seq(along=rectInfo),SIMPLIFY=FALSE)) ## add rectangle to original plot of points p+layer(geom=polygon,data=x,mapping=aes(x=x,y=y),facets=f~.) # will print the graphics on my windows() device Though lattice does seem to emphasize the 'chart type' approach to graphing, in a way I see that it provides a similar flexibility - just that the specifications for each element are contained in functions and objects that are ultimately invoked by a high-level/higher-order function, instead of being combined in the linear fashion of ggplot2. ST --- Deepayan Sarkar [EMAIL PROTECTED] wrote: On 7/11/07, hadley wickham [EMAIL PROTECTED] wrote: A question/comment: I have usually found that the subscripts argument is what I need when passing *external* information into the panel function, for example, when I wish to add results from a fit done external to the trellis call. Fits[subscripts] gives me the fits (or whatever) I want to plot for each panel. It is not clear to me how the panel layout information from panel.number(), etc. would be helpful here instead. Am I correct? -- or is there a smarter way to do this that I've missed? This is one of things that I think ggplot does better - it's much easier to plot multiple data sources. I don't have many examples of this yet, but the final example on http://had.co.nz/ggplot2/geom_abline.html illustrates the basic idea. That's probably true. The Trellis approach is to define a plot by data source + type of plot, whereas the ggplot approach (if I understand correctly) is to create a specification for the display (incrementally?) and then render it. Since the specification can be very general, the approach is very flexible. The downside is that you need to learn the language. On a philosophical note, I think the apparent limitations of Trellis in some (not all) cases is just due to the artificial importance given to data frames as the one true container for data. Now that we have proper multiple dispatch in S4, we can write methods that behave like traditional Trellis calls but work with more complex data structures. We have tried this in one bioconductor package (flowViz) with encouraging results. -Deepayan __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Pinpoint customers who are looking for what you sell. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Drawing rectangles in multiple panels
On 7/12/07, Deepayan Sarkar [EMAIL PROTECTED] wrote: On 7/11/07, hadley wickham [EMAIL PROTECTED] wrote: A question/comment: I have usually found that the subscripts argument is what I need when passing *external* information into the panel function, for example, when I wish to add results from a fit done external to the trellis call. Fits[subscripts] gives me the fits (or whatever) I want to plot for each panel. It is not clear to me how the panel layout information from panel.number(), etc. would be helpful here instead. Am I correct? -- or is there a smarter way to do this that I've missed? This is one of things that I think ggplot does better - it's much easier to plot multiple data sources. I don't have many examples of this yet, but the final example on http://had.co.nz/ggplot2/geom_abline.html illustrates the basic idea. That's probably true. The Trellis approach is to define a plot by data source + type of plot, whereas the ggplot approach (if I understand correctly) is to create a specification for the display (incrementally?) and then render it. Since the specification can be very general, the approach is very flexible. The downside is that you need to learn the language. Yes, that's right. ggplot basically decomposes type of plot into statistical transformation (stat) + geometric object and allows you to control each component separately. ggplot also explicitly includes the idea of layers (ie. one layer is a scatterplot and another layer is a loess smooth) and allows you to supply different datasets to different layers. On a philosophical note, I think the apparent limitations of Trellis in some (not all) cases is just due to the artificial importance given to data frames as the one true container for data. Now that we have proper multiple dispatch in S4, we can write methods that behave like traditional Trellis calls but work with more complex data structures. We have tried this in one bioconductor package (flowViz) with encouraging results. That's one area which I haven't thought much about. ggplot is very data.frame centric and it's not yet clear to me how plotting a linear model (say) would fit into the grammar. Hadley __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Drawing rectangles in multiple panels
On 7/12/07, Stephen Tucker [EMAIL PROTECTED] wrote: Not that Trellis/lattice was entirely easy to learn at first. :) I've been playing around with ggplot2 and there is a plot()-like wrapper for building a quick plot [incidentally, called qplot()], but otherwise it's my understanding that you superpose elements (incrementally) to build up to the graph you want. Here is the same plot in ggplot2: rectInfo - list(matrix(runif(4), 2, 2), matrix(runif(4), 2, 2), matrix(runif(4), 2, 2)) library(ggplot2) ggopt(grid.fill = white) # just my preference ## original plot of points p - qplot(x,y,data=data.frame(x=runif(30),y=runif(30),f=gl(3,30)),facets=f~.) # print(p) ## external data (rectangles) - in coordinates for geom_polygon x - do.call(rbind, mapply(function(.r,.f) data.frame(x=.r[c(1,1,2,2),1],y=.r[c(1,2,2,1),2],f=.f), .r=rectInfo,.f=seq(along=rectInfo),SIMPLIFY=FALSE)) ## add rectangle to original plot of points p+layer(geom=polygon,data=x,mapping=aes(x=x,y=y),facets=f~.) # will print the graphics on my windows() device You should be able to simplify this line to: p+geom_polygon(data=x) because all the other information is already contained in the plot object. Though lattice does seem to emphasize the 'chart type' approach to graphing, in a way I see that it provides a similar flexibility - just that the specifications for each element are contained in functions and objects that are ultimately invoked by a high-level/higher-order function, instead of being combined in the linear fashion of ggplot2. I tend to think in very data centric approach, where you first generate the data (in a data frame) and then you plot it. There is very little data creation/modification during the plotting itself - I think this is different to lattice, where you often do more data manipulation in the panel function itself. I don't think one is better or worse, just different. Hadley __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Drawing rectangles in multiple panels
On 7/12/07, Stephen Tucker [EMAIL PROTECTED] wrote: In the Trellis approach, another way (I like) to deal with multiple pieces of external data sources is to 'attach' them to panel functions through lexical closures. For instance... rectInfo - list(matrix(runif(4), 2, 2), matrix(runif(4), 2, 2), matrix(runif(4), 2, 2)) panel.qrect - function(rect.info) { function(x, y, ...) { ri - rect.info[[packet.number()]] panel.rect(ri[1, 1], ri[1, 2], ri[2, 1], ri[2, 2], col = grey86, border = NA) panel.xyplot(x, y, ...) } } xyplot(runif(30) ~ runif(30) | gl(3, 10), panel = panel.qrect(rectInfo)) ...which may or may not be more convenient than passing rectInfo (and perhaps other objects if desired) explicitly as an argument to xyplot(). This is an interesting approach. The one problem I see with it is that if you change the trellising specification, you have to change your rectInfo datastructure. I guess we're missing the code that actually generates rectInfo in the first place, so maybe in practice it's not such a big problem. Hadley __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Subsetting problem
I need to perform the Exact Wilcoxon Mann-Whitney on a subset of my database. Assuming that IPPO is my data frame and IPPOBIS is the subset my variable still have 3 different levels and the function wilcox_test (package coin) does not accept it. I do not know how to overcome this problem. ippo - c(rep(A,10),rep(B,10),rep(C,10)) ippo2 - c(rnorm(10,0,1),rnorm(10,10,10),rnorm(10,10,10)) IPPO - data.frame(ippo,ippo2) IPPOBIS - IPPO[IPPO$ippo == A | IPPO$ippo == B,] wilcox_test(ippo2 ~ ippo,data=IPPOBIS,distribution=exact()) Error in check(itp) : 'object' does not represent a two sample problem levels(IPPOBIS$ippo) [1] A B C Massimo Cressoni __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] make error R-5.1 on sun solaris
You are asked for a C99 compiler and configure normally finds one: mixing declarations and code is valid C99. Unless something has been done with environment variables (e.g. in config.site) this gcc is very old. configure should come up with 'gcc -std=gnu99'. Re-ordering the code will help (but it may need to go below the next #ifdef block too), but a gcc update would be a very good idea as gcc on Sparc has been buggy up to about 3.4.x. Note too that --without-iconv is undesirable and should not be necessary as libiconv can be installed as a preload on Solaris (8 and 10, so presumably also 9). On Thu, 12 Jul 2007, Peter Dalgaard wrote: Dan Powers wrote: I hope this is enough information to determine the problem. Thanks in advance for any help. Configure goes ok (I think) ./configure --prefix=$HOME --without-iconv R is now configured for sparc-sun-solaris2.9 Source directory: . Installation directory:/home/dpowers C compiler:gcc -g -O2 Fortran 77 compiler: f95 -g C++ compiler: g++ -g -O2 Fortran 90/95 compiler:f95 -g Obj-C compiler: -g -O2 Interfaces supported: X11 External libraries:readline Additional capabilities: NLS Options enabled: shared BLAS, R profiling, Java Recommended packages: yes Make ends after the gcc.. make . . . gcc -I. -I../../src/include -I../../src/include -I/usr/openwin/include -I/usr/local/include -DHAVE_CONFIG_H -g -O2 -c system.c -o system.o system.c: In function `Rf_initialize_R': system.c:144: parse error before `char' system.c:216: `localedir' undeclared (first use in this function) system.c:216: (Each undeclared identifier is reported only once system.c:216: for each function it appears in.) *** Error code 1 make: Fatal error: Command failed for target `system.o' Current working directory /home/dpowers/R-2.5.1/src/unix *** Error code 1 make: Fatal error: Command failed for target `R' Current working directory /home/dpowers/R-2.5.1/src/unix *** Error code 1 make: Fatal error: Command failed for target `R' Current working directory /home/dpowers/R-2.5.1/src *** Error code 1 make: Fatal error: Command failed for target `R' I have tried setting localedir directly in configure options, but get the same error. Any ideas? Hmm, which version of gcc is this? The problem seems to be around line 144 which reads 140 Rstart Rp = rstart; 141 cmdlines[0] = '\0'; 142 143 #ifdef ENABLE_NLS 144 char localedir[PATH_MAX+20]; 145 #endif 146 147 #if defined(HAVE_SYS_RESOURCE_H) defined(HAVE_GETRLIMIT) 148 { 149 struct rlimit rlim; I seem to remember that it used to be non-kosher to mix declarations and ordinary code like that, but the current compiler doesn't seem to care (I do have #define ENABLE_NLS 1 in Rconfig.h, as I assume you do too). Could you perhaps try moving line 141 down below #endif? Thanks, Dan =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Daniel A. Powers, Ph.D. Department of Sociology University of Texas at Austin 1 University Station A1700 Austin, TX 78712-0118 phone: 512-232-6335 fax: 512-471-1748 [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] ggplot2 / reshape / Question on manipulating data
I'm an R newbie but recently discovered the ggplot2 and reshape packages which seem incredibly useful and much easier to use for a beginner. Using the data from the IMDB, I'm trying to see how the average movie rating varies by year. Here is what my data looks like: ratings - read.delim(groomed.list, header = TRUE, sep = |, comment.char = ) ratings - subset(ratings, VoteCount 100) head(ratings) Title Histogram VoteCount VoteMean Year 1!Huff (2004) (TV) 16 299 8.4 2004 8 'Allo 'Allo! (1982) 000125 829 8.6 1982 50 .hack//SIGN (2002) 001113 150 7.0 2002 561-800-Missing (2003) 000103 118 5.4 2003 66 Greatest Artists (2000) (mini) 00..16 110 7.8 2000 77 00 Scariest Movie (2004) (mini) 00..000115 256 8.6 2004 The above data is not aggregated. So after playing around with basic R functionality, I stumbled across the 'aggregate' function and was able to see the information in the manner I desired (average movie rating by year). byYear - aggregate(ratings$VoteMean, list(Year = ratings$Year), mean) plot(byYear) Having just discovered gglot2, I wanted to create the same graph but augment it with a color attribute based on the total number of votes in a year. So first I tried to see if I could reproduce the above: library(ggplot2) qplot(Year, x, byYear) This did not work as expected because the x-axis contained labels for each and every year making it impossible to read whereas the plot created with basic R had nice x-axis labels. How do I get 'qplot' to treat the x-axis in a similar manner to 'plot'? After playing around further, I was able to get 'qplot' to work in a manner similar to 'plot' with regards to the x-axis labels by using 'melt' and 'cast'. The 'qplot' now behaves correctly: mratings - melt(ratings, id = c(Title, Year), measure = c(VoteCount, VoteMean)) byYear2 - cast(mratings, Year ~ variable, mean, subset = variable == VoteMean) qplot(Year, VoteMean, data = byYear2) How do 'byYear' and 'byYear2' differ? I am trying to use 'typeof' but both seem to be lists. However, they are clearly different in some way because 'qplot' graphs them differently. Finally, I'd like to use a color attribute to 'qplot' to augment each point with a color based on the total number of votes for the year. Using attributes with 'qplot' seems simple, but I'm having a hard time grooming my data appropriately. I believe this requires aggregation by summing the VoteCount column. Is there a way to cast the data using different aggregation functions for various columns? In my case, I want the mean of the VoteMean column, and the sum of the VoteCount column. Then I want to produce a graph showing the average movie rating per year but with each point colored to reflect the total number of votes for that year. Any pointers? Thanks, Pete __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Subsetting problem
You have three levels of factor 'ippo' and data on two. That is not a two-sample problem, as the error message says. Try IPPOBIS$ippo - IPPOBIS$ippo[drop=TRUE] And please use an informative subject line (see the posting guide). On Thu, 12 Jul 2007, Cressoni, Massimo (NIH/NHLBI) [F] wrote: I need to perform the Exact Wilcoxon Mann-Whitney on a subset of my database. Assuming that IPPO is my data frame and IPPOBIS is the subset my variable still have 3 different levels and the function wilcox_test (package coin) does not accept it. I do not know how to overcome this problem. ippo - c(rep(A,10),rep(B,10),rep(C,10)) ippo2 - c(rnorm(10,0,1),rnorm(10,10,10),rnorm(10,10,10)) IPPO - data.frame(ippo,ippo2) IPPOBIS - IPPO[IPPO$ippo == A | IPPO$ippo == B,] wilcox_test(ippo2 ~ ippo,data=IPPOBIS,distribution=exact()) Error in check(itp) : 'object' does not represent a two sample problem levels(IPPOBIS$ippo) [1] A B C Massimo Cressoni __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.