Re: [R] can't understand R
Erin L. Leisz wrote: hi. i really need help using this program. computer language is a foreign language to me, and thus, i cannot make heads nor tails of the user manuals from the website. i need to locate step-by-step examples of simple problems such as graph f(x)+g(x) and f(g(x)) for the domain 0x2 and graph 2H(x), H(x)+1, H(x+1) i do know how to define the functions, but that's it. is there any help you could provide me? i would appreciate some help asap. thank you very much erin leisz If the manual An Introduction to R is not sufficient for you, what about reading a book, e.g. Peter Dalgaard's Introductory Statistics with R, Springer? Here you learn R along some basic statistical analyses. Uwe Ligges __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] bubble plots
Dear all, I'm used to draw bubble plots (gstat) to get a first view of my spatial marked data, but I can't find a way to label legend with marks after having replaced them by a numeric scale. I have a data frame with numeric coords and factors: names(data) [1] x y bloc sub inoc etat etc... levels(data$etat) [1] F Fi NF NFi I do: levels(data$etat) - c(0, 2, -2, 1, -1) data$etat - as.numeric(as.vector(data$etat)) C1 - subset(data, sub==C bloc==1, select=c(x, y, etat)) x11() lset(theme = col.whitebg()) bubble(C1, xcol = 1, ycol = 2, zcol = 3, xlab = , ylab = , main = Bloc 1 - Substrat C, key.entries = min(C1[, 3]):max(C1[, 3]), maxsize=3) I then obtain -2, -1, O, etc. as legend labels and cannot find argument making it possible to label my legend with initial marks: F1, NFi, etc. Besides, I also tried with image() but I faced the following error message, even though I ordered x- and y-coords: Error in image.default(C1$x, C1$y, C1$etat) : increasing x and y values expected I'd really appreciate if someone could give me either an option to label adequately bubble plots, or even to make me know if there is another - maybe better - way to plot spatial marked data. Jacques VESLOT CIRAD Réunion __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] can't understand R
Hello, Uwe Ligges schrieb: Erin L. Leisz wrote: If the manual An Introduction to R is not sufficient for you, what about reading a book, e.g. Peter Dalgaard's Introductory Statistics with R, Springer? Here you learn R along some basic statistical analyses. BTW, can anybody recommend a book on S/R and data mining? Thomas __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] unable to load shared library /home/hpc1367/runs/taper/taper.so
hpc1367 == hpc1367 [EMAIL PROTECTED] on Mon, 20 Sep 2004 18:18:58 -0400 writes: Why do you put the full filename in the e-mail subject? (We'd rather want to know your operating system, some kind of unix, probably Linux, seen from below) hpc1367 I am trying to load a .so file and get the hpc1367 following error message: dyn.load(taper.so,local=F) hpc1367 Error in dyn.load(x, as.logical(local), hpc1367 as.logical(now)) : unable to load shared library hpc1367 /home/hpc1367/runs/taper/taper.so: ld.so.1: hpc1367 /usr/local/lib/R/bin/R.bin: fatal: relocation hpc1367 error: file /home/hpc1367/runs/taper/taper.so: hpc1367 symbol f90_init: referenced symbol not found this is the crucial part ! hpc1367 also tried with default local=T and got same error. of course, since your fortran90 code needsf90_init() and that is evidently not linked into your taper.so but is provided by your Fortran90 hpc1367 my makefile (used to produce the .so) is included below hpc1367 F90S = taper.f90 hpc1367 OBJS = taper.o hpc1367 LIBS = -lsocket -lnsl -lintl hpc1367 OPT = -fast hpc1367 MODS = -M/opt/NAG/fnsol04dbl/nag_mod_dir hpc1367 NAG = /opt/NAG/fnsol04dbl/libnagfl90.a hpc1367 STATNAG = /opt/NAG/flsol20dal/libnag.a hpc1367 taper.o : taper.f90 hpc1367 f95 -c -dalign $(OPT) $(MODS) taper.f90 hpc1367 taper.so: taper.f90 hpc1367 f95 -o taper.so -G -dalign $(OPT) $(MODS) taper.f90 hpc1367 taper.x : taper.o hpc1367 f95 -o taper.x -dalign $(OPT) $(MODS) $(LIBS) taper.o $(OBJS) hpc1367 $(NAG) You must make sure that the system library which defines the f90_init() symbol is specified at link time, i.e., when taper.so is built. So I guess you need at least the $(LIBS) also in that line, probably also $(NAG). But we don't have your non-free NAG compilers and libraries and it's hard to find out from here. Use 'nm' and 'ldd' but probably rather ask someone at your institution who knows more about compilation and linking. hpc1367 please, help hpc1367 thank you very much __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] installing on alpha
Alexa == Alexa Sorant [EMAIL PROTECTED] on Mon, 20 Sep 2004 17:35:43 -0400 writes: Alexa I am trying to install R-1.9.1 on an alpha (Tru64 unix) platform. I had no Alexa problem previously installing R-1.4.0, but cannot get the new installation Alexa to work. First I ran configure specifying an alternate installation Alexa directory: Alexa ./configure --prefix=/appl/R-1.9.1 We even recommend using an alternate *build* directory: mkdir Rbuild cdRbuild ../Rsrc/configure and indeed, since R 2.0.0 is beta now (which means testing and almost stable, not unstable!), we'd appreciate (and feel more willing to support you) if you'd tried to install R-2.0.0 directly. Alexa then ran make (note that this is NOT gmake, which Alexa may make the difference). The error message I get is: Alexa Make: cannot open /share/make/vars.mk. Stop. Alexa The file share/make/vars.mk is there and readable, Alexa but share is a subdirectory of the main installation Alexa directory. Upon examining Makefile and associated Alexa files, I found a statement in Makeconf include Alexa $(top_srcdir)/share/make/vars.mk which I suspected Alexa was the problem -- perhaps the definition of Alexa top_srcdir not is carrying over from the Makefile Alexa that references Makeconf. yes, that seems clear. If your 'make' is not supporting such 'include' statements in Makefiles, you definitely must use GNU make (seemingly called 'gmake' on alpha) ! Did you look at the INSTALL file, or even consult the R Administration and Installation manual (available from R-projec.org as well) ? Alexa (One reason I suspect this statement is that it is new Alexa since the version of R that I installed easily, as Alexa well as the early reference to this particular file.) Alexa I have tried a bunch of ways of modifying this Alexa statement, including an explicit path in this Alexa statement. I think that did work, but a similar Alexa problem then occurred at a later point that I haven't Alexa yet located. Alexa Has anyone gotten this installation to work on this platform? Alexa Alexa J. M. Sorant Alexa NIH/NHGRI Alexa 333 Cassell Drive Suite 2000 Alexa Baltimore, MD 21224 Alexa [EMAIL PROTECTED] Alexa (410) 550-7512 voice Alexa (410) 550-7513 fax __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] plot.Map with pattern instead of colors
Dear Roger, thank you for your valuable hints regarding plot.polylist, Map2poly and RColorBrewer( , Greys ). Adding pattern to some shapes with plot( Map2poly( x ), density = myDensities, add = TRUE ) works great! Best wishes, Arne On Monday 20 September 2004 14:20, Roger Bivand wrote: On Mon, 20 Sep 2004, Arne Henningsen wrote: Hi, I am plotting shapefiles with plot.Map (package maptools). So far I use different colors for the shapes depending on the a value that belongs to the shape. Now, I need to produce maps only in black, gray and white for publication. Is it possible to fill shapes with pattern (e.g. hatched) instead of colors? Not in plot.Map(). This is supported if you convert the shapefile polygons to a polylist object (Map2poly()), then use plot.polylist() - but this may not be your choice. Within plot.Map, I would suggest using the RColorBrewer package and the sequential Greys palette, which differentiates five grey colours very well for most media. If you run this after example(plot.Map), it should illustrate a solution: library(RColorBrewer) pal - brewer.pal(5, Greys) fgs - pal[findInterval(x$att.data$BIR74, res$breaks, all.inside=TRUE)] plot(x, fg=fgs) for the default quantile breaks. Details of a simplified example: I want to show the percentage change of a variable in a map: e.g. following levels a) +10 b) +5% to +10% c) -5% to +5% d) -10% to -5% e) -10% Now I have following colors a) red b) orange c) white d) greenyellow e) green3 Printing this on a monochrome printer is - of course - stupid, because levels a) and e) as well as b) and d) have approximately the same grayscale. I could fill a) = black and e) = white, and everything inbetween with an appropriate grayscale, but I prefer to have areas with no change to appear white rather than medium gray. Therefore, I thought of doing following: a) black b) gray c) white d) hatched with thin lines e) hatched with thick lines (or double-hatched) Any hints and ideas are welcome! (BTW: I use R 1.9.1 on SuSE Linux 9.0) Thanks, Arne -- Arne Henningsen Department of Agricultural Economics University of Kiel Olshausenstr. 40 D-24098 Kiel (Germany) Tel: +49-431-880 4445 Fax: +49-431-880 1397 [EMAIL PROTECTED] http://www.uni-kiel.de/agrarpol/ahenningsen/ __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] S/R and data mining (was can't understand R)
Hi Thomas, @Book{hastie.et.al:01, author= {T. Hastie and R. Tibshirani and J. Friedman}, address = {New York}, publisher = {Springer-Verlag}, title = {The Elements of Statistical Learning: Data Mining, Inference and Prediction}, year = {2001} } might be helpful. Best, Dimitris Dimitris Rizopoulos Ph.D. Student Biostatistical Centre School of Public Health Catholic University of Leuven Address: Kapucijnenvoer 35, Leuven, Belgium Tel: +32/16/396887 Fax: +32/16/337015 Web: http://www.med.kuleuven.ac.be/biostat/ http://www.student.kuleuven.ac.be/~m0390867/dimitris.htm - Original Message - From: Thomas Schönhoff [EMAIL PROTECTED] To: R User-Liste [EMAIL PROTECTED] Sent: Tuesday, September 21, 2004 9:34 AM Subject: Re: [R] can't understand R Hello, Uwe Ligges schrieb: Erin L. Leisz wrote: If the manual An Introduction to R is not sufficient for you, what about reading a book, e.g. Peter Dalgaard's Introductory Statistics with R, Springer? Here you learn R along some basic statistical analyses. BTW, can anybody recommend a book on S/R and data mining? Thomas __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] compilation failed for package
On Mon, 20 Sep 2004 [EMAIL PROTECTED] wrote: Hello, I am new to R on Linux (and to LINUX), running a 1.9.1 on a Linux MDK10.0. When updating or installing different packages, I get messages like this (under), the package doesn't install. Can anyone help me? I know this is all to little information, but I don't know were to start. I think there is something wrong in how or were the C++ or fortran compilator is installed?? Thanks, Geir S. After some offline exchanges, the problem has been resolved. This Mandrake 10.0 installation had R installed from binary, had gcc 3.4.1, but (most likely) binutils 2.14 that did not recognise the new --as-needed flag. For this reason, update.packages() - this was the survival case mentioned in the post - and install.packages() failed on compilation. The fact that such a build system is broken is demonstrated by: $ cat EOF hello.c #include stdio.h int main(void) { printf(Hello World!\n); exit(0); } EOF $ gcc -o hello hello.c /usr/bin/ld: unrecognized option '--as-needed' /usr/bin/ld: use the --help option for usage information collect2: ld returned 1 exit status Resolution by updating binutils or installing older gcc-* RPMs instead of 3.4.1. gcc -shared -L/usr/local/lib -o survival.so agexact.o agfit2.o agfit3.o agfit5.o agfit_null.o agmart2.o agmart.o agscore.o agsurv1.o agsurv2.o agsurv3.o char_date.o chinv2.o chinv3.o cholesky2.o cholesky3.o chsolve2.o chsolve3.o coxdetail.o coxfit2.o coxfit5.o coxmart.o coxph_wtest.o cox_Rcallback.o coxscho.o coxscore.o dmatrix.o doloop.o pyears1.o pyears2.o pyears3.o pystep.o surv_callback.o survdiff2.o survfit2.o survfit3.o survindex2.o survindex3.o survreg2.o survreg3.o survreg4.o survreg5.o /usr/bin/ld: unrecognized option '--as-needed' /usr/bin/ld: use the --help option for usage information collect2: ld returned 1 exit status make: *** [survival.so] Error 1 ERROR: compilation failed for package 'survival' ** Removing '/usr/lib/R/library/survival' ** Restoring previous '/usr/lib/R/library/survival' __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Roger Bivand Economic Geography Section, Department of Economics, Norwegian School of Economics and Business Administration, Breiviksveien 40, N-5045 Bergen, Norway. voice: +47 55 95 93 55; fax +47 55 95 93 93 e-mail: [EMAIL PROTECTED] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] adding function to a Package
Stephen Henderson wrote: As a shortcut I have previously added my own functions to a Package (ipred) under R-1.8.1, changing the NAMESPACE file to accomodate this then reinstalling. This doesn't now seem possible in R-1.9.0 is this correct?-- I am getting an error saying: files NAMESPACE, R/ipred have the wrong MD5 checksums after I try to reinstall (which I can understand). Is there a quick way round this avoiding having to build my own Package? Looks like you have added to the binary rather than the soource package. Please change the source package instaed and re-install. BTW: It is a better idea to put the stuff in your own package, because it will be hard for you to go with each update of ipred, since you have to re-apply your patches again and again ... Uwe Ligges Thanks Stephen ** This email and any files transmitted with it are confidentia...{{dropped}} __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] S/R and data mining (was can't understand R)
Hi Thomas, see these papers or books (some are available on the web): Diego Kuonen, Introduction au data mining avec R : vers la reconquête du `knowledge discovery in databases' par les statisticiens. Bulletin of the Swiss Statistical Society, 40:3-7, 2001. Consultabile allindirizzo web: http://www.statoo.com/en/publications/2001.R.SSS.40/ Diego Kuonen and Reinhard Furrer, Data mining avec R dans un monde libre. Flash Informatique Spécial Été, pages 45-50, sep 2001. Consultabile allindirizzo web: http://sawww.epfl.ch/SIC/SA/publications/FI01/fi-sp-1/sp-1-page45.html Brian D. Ripley, Datamining: Large Databases and Methods, in Proceedings of useR! 2004 - The R User Conference, maggio 2004 Consultabile allindirizzo web: http://www.ci.tuwien.ac.at/Conferences/useR-2004/Keynotes/Ripley.pdf Brian D. Ripley, Using Databases with R, R News, Gennaio 2001, pagg. 18-20 Consultabile allindirizzo web: http://cran.r-project.org/doc/Rnews/Rnews_2001-1.pdf B. D. Ripley, R. M. Ripley, Applications of R Clients and Servers in Proceedings of the Distributed Statistical Computing 2001 Workshop, 2001, Vienna University of Technology. Consultabile allindirizzo web: http://www.ci.tuwien.ac.at/Conferences/DSC-2001/Proceedings/Ripley.pdf Torsten Hothorn, David A. James, Brian D. Ripley, R/S Interfaces to Databases in Proceedings of the Distributed Statistical Computing 2001 Workshop, 2001,Vienna University of Technology. Consultabile allindirizzo web: http://www.ci.tuwien.ac.at/Conferences/DSC-2001/Proceedings/HothornJamesRipley.pdf Luís Torgo, Data Mining with R. Learning by case studies, Maggio 2003 Consultabile allindirizzo web: http://www.liacc.up.pt/~ltorgo/DataMiningWithR/ Trevor Hastie , Robert Tibshirani, Jerome Friedman, The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2001, Springer-Verlag. http://www-stat.stanford.edu/~tibs/ElemStatLearn/ I'm find too for some R application for DM. Best Vito Thomas Schönhoff wrote: Hello, Uwe Ligges schrieb: Erin L. Leisz wrote: If the manual An Introduction to R is not sufficient for you, what about reading a book, e.g. Peter Dalgaard's Introductory Statistics with R, Springer? Here you learn R along some basic statistical analyses. BTW, can anybody recommend a book on S/R and data mining? Thomas = Diventare costruttori di soluzioni Visitate il portale http://www.modugno.it/ e in particolare la sezione su Palese http://www.modugno.it/archivio/cat_palese.shtml ___ http://it.seriea.fantasysports.yahoo.com/ __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] can't understand R
Dear Erin, On Tuesday 21 September 2004 06:10, Erin L. Leisz wrote: hi. i really need help using this program. computer language is a foreign language to me, and thus, i cannot make heads nor tails of the user manuals from the website. i need to locate step-by-step examples of simple If you plan to use R more than once, I think you probably want to get used to using the manuals (starting with An introduction to R, and maybe some of the other intro material available from the R web site). problems such as graph f(x)+g(x) and f(g(x)) for the domain 0x2 and graph 2H(x), H(x)+1, H(x+1) i do know how to define the functions, but that's it. is there any help you could provide me? i would appreciate some help asap. thank you very much For this particular case of plotting f(x), you can take a look at the function curve (do: ?curve at the R prompt). Hope this helps. R. erin leisz __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Ramón Díaz-Uriarte Bioinformatics Unit Centro Nacional de Investigaciones Oncológicas (CNIO) (Spanish National Cancer Center) Melchor Fernández Almagro, 3 28029 Madrid (Spain) Fax: +-34-91-224-6972 Phone: +-34-91-224-6900 http://ligarto.org/rdiaz PGP KeyID: 0xE89B3462 (http://ligarto.org/rdiaz/0xE89B3462.asc) __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] can't understand R
On Tue, Sep 21, 2004 at 09:34:23AM +0200, Thomas Schönhoff wrote: BTW, can anybody recommend a book on S/R and data mining? Try this one, it's not finished but yet available for free: ``Data Mining with R - learning by case studies'' by Luís Torgo; http://www.liacc.up.pt/~ltorgo/DataMiningWithR/. Regards Sven C. Koehler __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
SV: [R] compilation failed for package
Thanks to Roger Bivand, my problem was solved. binutils 2.14 was updatet to binutils 2.15, as indicated under. Geir Helge Systad Norwegian Institute for Nature Research Arctic Ecology, Polarmiljsenteret, N-9296 Troms, Norway Adress aug-04 to feb-05: Albert Pettersonsvei 13, N-5750 Odda fax +47 85 03 82 14 phone+47 53 64 24 94 mobil+47 91 63 70 55 E-mail: [EMAIL PROTECTED] Internett: http://www.ninaniku.no -Opprinnelig melding- Fra:Roger Bivand [mailto:[EMAIL PROTECTED] Sendt: ti 21.09.2004 10:20 Til:Systad, Geir Kopi: [EMAIL PROTECTED] Emne: Re: [R] compilation failed for package On Mon, 20 Sep 2004 [EMAIL PROTECTED] wrote: Hello, I am new to R on Linux (and to LINUX), running a 1.9.1 on a Linux MDK10.0. When updating or installing different packages, I get messages like this (under), the package doesn't install. Can anyone help me? I know this is all to little information, but I don't know were to start. I think there is something wrong in how or were the C++ or fortran compilator is installed?? Thanks, Geir S. After some offline exchanges, the problem has been resolved. This Mandrake 10.0 installation had R installed from binary, had gcc 3.4.1, but (most likely) binutils 2.14 that did not recognise the new --as-needed flag. For this reason, update.packages() - this was the survival case mentioned in the post - and install.packages() failed on compilation. The fact that such a build system is broken is demonstrated by: $ cat EOF hello.c #include stdio.h int main(void) { printf(Hello World!\n); exit(0); } EOF $ gcc -o hello hello.c /usr/bin/ld: unrecognized option '--as-needed' /usr/bin/ld: use the --help option for usage information collect2: ld returned 1 exit status Resolution by updating binutils or installing older gcc-* RPMs instead of 3.4.1. gcc -shared -L/usr/local/lib -o survival.so agexact.o agfit2.o agfit3.o agfit5.o agfit_null.o agmart2.o agmart.o agscore.o agsurv1.o agsurv2.o agsurv3.o char_date.o chinv2.o chinv3.o cholesky2.o cholesky3.o chsolve2.o chsolve3.o coxdetail.o coxfit2.o coxfit5.o coxmart.o coxph_wtest.o cox_Rcallback.o coxscho.o coxscore.o dmatrix.o doloop.o pyears1.o pyears2.o pyears3.o pystep.o surv_callback.o survdiff2.o survfit2.o survfit3.o survindex2.o survindex3.o survreg2.o survreg3.o survreg4.o survreg5.o /usr/bin/ld: unrecognized option '--as-needed' /usr/bin/ld: use the --help option for usage information collect2: ld returned 1 exit status make: *** [survival.so] Error 1 ERROR: compilation failed for package 'survival' ** Removing '/usr/lib/R/library/survival' ** Restoring previous '/usr/lib/R/library/survival' __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Roger Bivand Economic Geography Section, Department of Economics, Norwegian School of Economics and Business Administration, Breiviksveien 40, N-5045 Bergen, Norway. voice: +47 55 95 93 55; fax +47 55 95 93 93 e-mail: [EMAIL PROTECTED] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] S/R and data mining (was can't understand R)
Thanks Dimitri and Vito, I'll soon have a look at your recommendations (except the mentioned book which might take some time to get hands at! Thomas __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] lda predict
Dear R-helpers, I have a model created by lda, and I would like to use this model to make predictions for new or old data. The catch is, I want to do this without using the predict function, i.e. only using information directly from the foo.lda object to create my posterior probabilities. In anticipation of likely responses, I will be brushing up my lda knowledge using the given references when I have time, but am being hassled for an answer asap! Cheers, Rob. [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] GARCH model query
Hello, I am trying to find an easy way to estimate the following: y = Function(x) + lag(x,1) + garch_error_component E.g. estimate the mean component first and then use garch from tseries on the residuals from step 1. Best Adrian Any clues? Best regards, Costas --- __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] lme RE variance computation
As I understand it lme (in R v1.9.x) estimates random effect variances on a log scale, constraining them to be positive. Whilst this seems sensible, it does lead to apparently biased estimates if the variance is actually zero - which makes our simulation results look strange. Whilst we need to think a bit deeper about it - I still haven't got my head around what a negative variance could mean - does anyone know a way to take away the contraint and allowing zero or negative variances? Steve. Dr Steve Roberts [EMAIL PROTECTED] Senior Lecturer in Medical Statistics, CMMCH NHS Trust and University of Manchester Biostatistics Group, 0161 275 5192 / 0161 276 5785 __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] lme RE variance computation
Hi Steve, Estimation problems for the variance components in linear mixed models are usually occur for two reasons: 1. Due to model misspecification, i.e., using years instead of decades may show no variability in the slopes 2. Because the data do not support the assumptions of the linear mixed model (i.e., positive definite covariance matrix for the random-effects = increasing variance with time). These may cause zero or even negative variance components. For more info you could take a look at Verbeke and Molenberghs (2000, Section 5.6) and Searle, Casella and McCullogh (1992, Section 3.5). I don't know the exact formulation you are using, but maybe you could consider an analogue of you model using gls, i.e., lme(..., random=~1|id) gls(..., corr=corCompSymm(form=~1|id)) The references mentioned above are: @Book{verbeke.molenberghs:00, author= {G. Verbeke and G. Molenberghs}, title = {Linear Mixed Models for Longitudinal Data}, year = {2000}, address = {New York}, publisher = {Springer-Verlag} } @Book{searle.et.al:92, author= {S. Searle and G. Cassela and C. McCulloch}, title = {Variance Components}, year = {1992}, address = {New York}, publisher = {Wiley} } I hope it helps. Best, Dimitris Dimitris Rizopoulos Ph.D. Student Biostatistical Centre School of Public Health Catholic University of Leuven Address: Kapucijnenvoer 35, Leuven, Belgium Tel: +32/16/396887 Fax: +32/16/337015 Web: http://www.med.kuleuven.ac.be/biostat/ http://www.student.kuleuven.ac.be/~m0390867/dimitris.htm - Original Message - From: Steve Roberts [EMAIL PROTECTED] To: [EMAIL PROTECTED] Sent: Tuesday, September 21, 2004 1:38 PM Subject: [R] lme RE variance computation As I understand it lme (in R v1.9.x) estimates random effect variances on a log scale, constraining them to be positive. Whilst this seems sensible, it does lead to apparently biased estimates if the variance is actually zero - which makes our simulation results look strange. Whilst we need to think a bit deeper about it - I still haven't got my head around what a negative variance could mean - does anyone know a way to take away the contraint and allowing zero or negative variances? Steve. Dr Steve Roberts [EMAIL PROTECTED] Senior Lecturer in Medical Statistics, CMMCH NHS Trust and University of Manchester Biostatistics Group, 0161 275 5192 / 0161 276 5785 __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] How to plot residuals vs. fitted of trgls object (spatial::surf.gls)
Hi! Would like to to make a plot of residuals vs. fitted and Y vs. predicted values from an object of class trgls as returned by spatial::surf.gls directly. (without calling spatial::predict.trgls) Is it possible? on which list components should I look at? /E __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] How to call a R function from C or C++ ?
Have you some simple examples ? [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] How to call a R function from C or C++ ?
Bruno Nogent wrote: Have you some simple examples ? See the manual Writing R Extensions. Uwe Ligges [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Bootstrap ICC estimate with nested data
I would appreciate some thoughts on using the bootstrap functions in the library bootstrap to estimate confidence intervals of ICC values calculated in lme. In lme, the ICC is calculated as tau/(tau+sigma-squared). So, for instance the ICC in the following example is 0.116: tmod-lme(CINISMO~1,random=~1|IDGRUP,data=TDAT) VarCorr(tmod) IDGRUP = pdLogChol(1) Variance StdDev (Intercept) 0.1829931 0.42 Residual1.3907732 1.179310 0.18299/(0.18299+1.39077) [1] 0.1162757 Using the bootstrap library, I can set up theta to do the ICC as follows: theta-function(x,DATA){tmod-lme(CINISMO~1,random=~1|IDGRUP,data=DATA[x,]) OUT-as.numeric(VarCorr(tmod)[[1]])/(as.numeric(VarCorr(tmod)[[1]])+as.numer ic(VarCorr(tmod)[[2]])) return(OUT)} Finally, I can run the bootstrap-t confidence limit function (or other functions) as follows: bootout-boott(1:nrow(TDAT),100,theta,data=TDAT) This seems to work, but the estimates also seem strange. For intance, the observed ICC value is larger than the 95% confidence intervals provided by the bootstrap. It occurs to me that the results might be strange because the sampling with replacement is being done without regard to group membership. That is, I might select individual 1 from group 1 10 times (even though in the sample the group only has 5 members), and I might not select any individuals from group 2. My fundamental question is: What are people's thoughts about using bootstaping in nested data? Does one have to sample with replacement taking into consideration the group structure in the data? If so, any suggestions on how to do this? Paul Bliese US Army Medical Research Unit - Europe __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] bubble plots
Jacques, please try the following: z=factor(c('a', 'b','c')) z [1] a b c Levels: a b c x = c(1,2,3) y = c(3,2,1) xyplot(y~x,groups=z,auto.key=T,asp=diff(range(y))/diff(range(x))) the asp argument takes care of one unit in x being equal to one unit in y. If you add e.g. pch=z to vary symbols, auto.key will have to be replaced by the more complex key argument, see ?xyplot Color and symbol type are better to distinguish factor variables than size is, so I'd recommend using xyplot rather than bubble (which uses xyplot to plot bubbles) -- Edzer __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Bootstrap ICC estimate with nested data
Paul, I think that you should account for the group structure. My reading of Davison and Hinkley Bootstrap Methods and their Application (1997, p. 100) suggests that for balanced data structures with more than, say, 10 clusters, one should apply the bootstrap to the clusters, but not within the clusters. They provide some further notes that you might find useful. I hope that this helps. Andrew On Tue, Sep 21, 2004 at 03:41:29PM +0200, Bliese, Paul D MAJ USAMH wrote: I would appreciate some thoughts on using the bootstrap functions in the library bootstrap to estimate confidence intervals of ICC values calculated in lme. In lme, the ICC is calculated as tau/(tau+sigma-squared). So, for instance the ICC in the following example is 0.116: tmod-lme(CINISMO~1,random=~1|IDGRUP,data=TDAT) VarCorr(tmod) IDGRUP = pdLogChol(1) Variance StdDev (Intercept) 0.1829931 0.42 Residual1.3907732 1.179310 0.18299/(0.18299+1.39077) [1] 0.1162757 Using the bootstrap library, I can set up theta to do the ICC as follows: theta-function(x,DATA){tmod-lme(CINISMO~1,random=~1|IDGRUP,data=DATA[x,]) OUT-as.numeric(VarCorr(tmod)[[1]])/(as.numeric(VarCorr(tmod)[[1]])+as.numer ic(VarCorr(tmod)[[2]])) return(OUT)} Finally, I can run the bootstrap-t confidence limit function (or other functions) as follows: bootout-boott(1:nrow(TDAT),100,theta,data=TDAT) This seems to work, but the estimates also seem strange. For intance, the observed ICC value is larger than the 95% confidence intervals provided by the bootstrap. It occurs to me that the results might be strange because the sampling with replacement is being done without regard to group membership. That is, I might select individual 1 from group 1 10 times (even though in the sample the group only has 5 members), and I might not select any individuals from group 2. My fundamental question is: What are people's thoughts about using bootstaping in nested data? Does one have to sample with replacement taking into consideration the group structure in the data? If so, any suggestions on how to do this? Paul Bliese US Army Medical Research Unit - Europe __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Andrew Robinson Ph: 208 885 7115 Department of Forest Resources Fa: 208 885 6226 University of Idaho E : [EMAIL PROTECTED] PO Box 441133W : http://www.uidaho.edu/~andrewr Moscow ID 83843 Or: http://www.biometrics.uidaho.edu No statement above necessarily represents my employer's opinion. - End forwarded message - -- Andrew Robinson Ph: 208 885 7115 Department of Forest Resources Fa: 208 885 6226 University of Idaho E : [EMAIL PROTECTED] PO Box 441133W : http://www.uidaho.edu/~andrewr Moscow ID 83843 Or: http://www.biometrics.uidaho.edu No statement above necessarily represents my employer's opinion. __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] how to do jaccknife validation with R
i can only find jackknife validation after boot. but this may be not the things i want. __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] R 1.9.1 Fails to Start on WinXP SP2
I have two computers both of which are running Windows XP SP2. R 1.9.1 runs just fine on one but not the other. Both installations of R came from the same installation package. When I click on the desktop icon an hourglass shows up then disappears. Similarly, if I start RGUI the process shows up in task-manager then abruptly disappears. Interestingly, I can access R in Excel via the D(COM) Server on both computers. Any ideas? Thanks, Scott __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] bubble plots
On Tuesday 21 September 2004 08:41, Edzer Pebesma wrote: Jacques, please try the following: z=factor(c('a', 'b','c')) z [1] a b c Levels: a b c x = c(1,2,3) y = c(3,2,1) xyplot(y~x,groups=z,auto.key=T,asp=diff(range(y))/diff(range(x))) the asp argument takes care of one unit in x being equal to one unit in y. FWIW, In R 2.0.0 (currently beta), you can also do this with aspect=iso. If you add e.g. pch=z to vary symbols, auto.key will have to be replaced by the more complex key argument, see ?xyplot An alternative would be to change the settings (which is the only way the panel function and auto.key can share information). Temporary changes in settings can be attached to a trellis object as follows: xyplot(y~x,groups=z,auto.key=T,asp=diff(range(y))/diff(range(x)), par.settings = list(superpose.symbol = list(cex = 2, pch = levels(z Deepayan __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] DSE: covariance of white noise
Hi R-Community, I estimated a VARMA model with bft in dse1 without input: A(L)yt = B(L)et I got the auto-regressive polynomial array A and the moving-average polynomial array B, but how can I access the covariances of the white noise et (disturbance vector), e.g. for simulation? Much thanks in advance, Hagen Schmoeller -- Dipl.-Ing. Hagen K. Schmöller Leiter Forschungsgruppe Stromerzeugung und -handel Institut für Elektrische Anlagen und Energiewirtschaft, RWTH Aachen Schinkelstraße 6, D-52056 Aachen, Germany Tel.: +49 (0)241 80-96734 Fax : +49 (0)241 80-92197 [EMAIL PROTECTED] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] png problem
Dear R-users, I have a small problem with the function png(), when used with the argument colortype=pseudo.cube. png(toto.png, colortype=pseudo.cube) image(matrix(rnorm(1), 100, 100)) dev.off() R is blocked at the last command (R does not print any prompt after the last command). Nothing is written in the file (Gimp indicates that the file is corrupted). However, png(toto.png) image(matrix(rnorm(1), 100, 100)) dev.off() works fine. I tried: options(X11colortype = pseudo.cube) png(toto.png) image(matrix(rnorm(1), 100, 100)) dev.off() But, here again, R is blocked. I tried to replace dev.off() by graphics.off(), but this does not resolve the problem. The problem does not occurs when the function X11() is used instead of the function png(). I searched through the mail archive, the FAQ, on google, but I did not found any solution to this problem. On the help page on the function png(), it is indicated that The colour handling will be that of the 'X11' device in use. I never used these functions before, but maybe png() is not suitable with colortype=pseudo.cube ? Can you tell me where I have missed something ? Thanks in Advance, Clément Calenge. version _ platform sparc-sun-solaris2.9 arch sparc os solaris2.9 system sparc, solaris2.9 status major1 minor9.1 year 2004 month06 day 21 language R [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] unable to install dse package
Error in file(file, r) : unable to open connection In addition: Warning message: cannot open file 'a:/project.txt' please help - [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] S-PLUS and R
Hello, How do I import data from Excel onto R? I am using S-Plus Envstat module. I need to use some of the data that already exist in the Envstat and put it into R to make graphs, find basic informations like mean, median, standard deviation, etc. I have been reading the help links, but I don't see anything in reference to this. Please help me! -Maher Lina __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] unable to install dse package
Error in file(file, r) : unable to open connection In addition: Warning message: cannot open file 'a:/project.txt' please help - [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] unable to install dse package
aarthi ireddy wrote: Error in file(file, r) : unable to open connection In addition: Warning message: cannot open file 'a:/project.txt' Error in question: Required information missing In addition: Warning message: Cannot reply if version of R, dse, and OS, as well as details about the way of installation are missing. Warning message: Name of e-mail sender is missing please help Uwe Ligges please help - [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] S-PLUS and R
[EMAIL PROTECTED] wrote: Hello, How do I import data from Excel onto R? I am using S-Plus Envstat module. I need to use some of the data that already exist in the Envstat and put it into R to make graphs, find basic informations like mean, median, standard deviation, etc. I have been reading the help links, but I don't see anything in reference to this. Please help me! See the R Data Import/Export Manual. Uwe Ligges -Maher Lina __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] unable to install dse package
aarthi ireddy [EMAIL PROTECTED] writes: Error in file(file, r) : unable to open connection In addition: Warning message: cannot open file 'a:/project.txt' Er, what were you trying to do and why would you be reading from the diskette? (Was there a previous message? I don't see it.) -- O__ Peter Dalgaard Blegdamsvej 3 c/ /'_ --- Dept. of Biostatistics 2200 Cph. N (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] S-PLUS and R
Save the excel file as a .csv file. The read about help(read.csv) Regards Wayne -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: 21 September 2004 16:41 To: [EMAIL PROTECTED] Subject: [R] S-PLUS and R Hello, How do I import data from Excel onto R? I am using S-Plus Envstat module. I need to use some of the data that already exist in the Envstat and put it into R to make graphs, find basic informations like mean, median, standard deviation, etc. I have been reading the help links, but I don't see anything in reference to this. Please help me! -Maher Lina __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html KSS Ltd Seventh Floor St James's Buildings 79 Oxford Street Manchester M1 6SS England Company Registration Number 2800886 Tel: +44 (0) 161 228 0040 Fax: +44 (0) 161 236 6305 mailto:[EMAIL PROTECTED]http://www.kssg.com The information in this Internet email is confidential and m...{{dropped}} __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] unable to install dse package
Error in file(file, r) : unable to open connection In addition: Warning message: cannot open file 'a:/project.txt' please help help(read.table) Hannu Kahra - [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] S-PLUS and R
Hi, a shortcut to import data from an Excel sheet is: In Excel 1) select cells including data 2) press CTRL+C (copy in clipboard) open a window starting R In R type read.delim(clipboard,header=TRUE) if the are headers in excel data or read.delim(clipboard,header=FALSE) if there are not headers See: http://www.sciviews.org/_rgui/projects/RDcom.html to use R in Excel. Best Vito You wrote: Hello, How do I import data from Excel onto R? I am using S-Plus Envstat module. I need to use some of the data that already exist in the Envstat and put it into R to make graphs, find basic informations like mean, median, standard deviation, etc. I have been reading the help links, but I don't see anything in reference to this. Please help me! -Maher Lina = Diventare costruttori di soluzioni Visitate il portale http://www.modugno.it/ e in particolare la sezione su Palese http://www.modugno.it/archivio/cat_palese.shtml ___ http://it.seriea.fantasysports.yahoo.com/ __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Problems with boot and optim
I am trying to bootstrap the parameters for a model that is estimated through the optim() function and find that when I make the call to boot, it runs but returns the exact same estimate for all of the bootstrap estimates. I managed to replicate the same problem using a glm() model but was able to fix it when I made a call to the variables as data frame by their exact names. But no matter how I refer to the variables in the het.fit function (see below) I get the same result. I could bootstrap it with the sample command and a loop, but then the analysis in the next step isn't as nice The code for the likelihood and the call to boot is below. I have tried numerous other permutations as well. I am using R 1.9.1 on Windows XP pro. Thanks Luke Keele #Define Likelihood lik.hetprobit -function(par, X, Y, Z){ #Pull Out Parameters Y - as.matrix(y) X - as.matrix(x) Z - as.matrix(z) K -ncol(X) M -ncol(Z) b - as.matrix(par[1:K]) gamma - as.matrix(par[K+1:M]) mu - (X%*%b) sd - exp(Z%*%gamma) mu.sd -(mu/sd) #Form Likelihood log.phi - pnorm(ifelse(Y == 0, -1, 1) * mu.sd, log.p = TRUE) 2 * sum(log.phi) } y - as.matrix(abhlth) x - as.matrix(reliten) ones = rep(1, nrow(x)) x = cbind(ones,x) z = as.matrix(abinfo) data.het - as.matrix(cbind(y,x,z)) het.fit - function(data){ mod - optim(c(1,0,0), lik.hetprobit, Y=data$y, X=data$x, Z=data$z, method=BFGS, control=list(fnscale=-1), hessian=T) c(mod$par) } case.fun - function(d,i) het.fit(d[i,]) het.case - boot(data.het, case.fun, R=50) Luke Keele Post-Doctoral Fellow in Quantitative Methods Nuffield College, Oxford University Oxford, UK [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Bootstrap ICC estimate with nested data
Have you considered simulate.lme in package nlme? This is not bootstrapping, but it's not obvious to me how to bootstrap with a complicated structure and get anything with a simple interpretation. More information on this is provided in Pinhiero and Bates (2000) Mixed-Effects Models in S and S-Plus (Springer). Bates is the primary architect of lme, nlme, etc. I have found this book quite valuable. hope this helps. spencer graves Andrew Robinson wrote: Paul, I think that you should account for the group structure. My reading of Davison and Hinkley Bootstrap Methods and their Application (1997, p. 100) suggests that for balanced data structures with more than, say, 10 clusters, one should apply the bootstrap to the clusters, but not within the clusters. They provide some further notes that you might find useful. I hope that this helps. Andrew On Tue, Sep 21, 2004 at 03:41:29PM +0200, Bliese, Paul D MAJ USAMH wrote: I would appreciate some thoughts on using the bootstrap functions in the library bootstrap to estimate confidence intervals of ICC values calculated in lme. In lme, the ICC is calculated as tau/(tau+sigma-squared). So, for instance the ICC in the following example is 0.116: tmod-lme(CINISMO~1,random=~1|IDGRUP,data=TDAT) VarCorr(tmod) IDGRUP = pdLogChol(1) Variance StdDev (Intercept) 0.1829931 0.42 Residual1.3907732 1.179310 0.18299/(0.18299+1.39077) [1] 0.1162757 Using the bootstrap library, I can set up theta to do the ICC as follows: theta-function(x,DATA){tmod-lme(CINISMO~1,random=~1|IDGRUP,data=DATA[x,]) OUT-as.numeric(VarCorr(tmod)[[1]])/(as.numeric(VarCorr(tmod)[[1]])+as.numer ic(VarCorr(tmod)[[2]])) return(OUT)} Finally, I can run the bootstrap-t confidence limit function (or other functions) as follows: bootout-boott(1:nrow(TDAT),100,theta,data=TDAT) This seems to work, but the estimates also seem strange. For intance, the observed ICC value is larger than the 95% confidence intervals provided by the bootstrap. It occurs to me that the results might be strange because the sampling with replacement is being done without regard to group membership. That is, I might select individual 1 from group 1 10 times (even though in the sample the group only has 5 members), and I might not select any individuals from group 2. My fundamental question is: What are people's thoughts about using bootstaping in nested data? Does one have to sample with replacement taking into consideration the group structure in the data? If so, any suggestions on how to do this? Paul Bliese US Army Medical Research Unit - Europe __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Spencer Graves, PhD, Senior Development Engineer O: (408)938-4420; mobile: (408)655-4567 __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] lme RE variance computation
It may help to explicitly distinguish between parameter estimates and the true but unknown and unknowable reality that is assumed to have generated the data. In that reality, all variance components are nonnegative. Generally inferior algorithms can compute negative variance components. These were quite useful prior to the advent of modern computers and software like the nlme package. I suggest you think very carefully about the problem you want to solve. I have found Pinheiro and Bates (2000) Mixed-Effects Models in S and S-Plus (Springer) quite helpful both in theory and in how to implement it. Bates is the primary architect of lme, nlme, etc., and this book provides basic documentation for that package. In particular, I would expect that simulate.lme could be quite useful if you have any doubts about any issue. hope this helps. spencer graves Dimitris Rizopoulos wrote: Hi Steve, Estimation problems for the variance components in linear mixed models are usually occur for two reasons: 1. Due to model misspecification, i.e., using years instead of decades may show no variability in the slopes 2. Because the data do not support the assumptions of the linear mixed model (i.e., positive definite covariance matrix for the random-effects = increasing variance with time). These may cause zero or even negative variance components. For more info you could take a look at Verbeke and Molenberghs (2000, Section 5.6) and Searle, Casella and McCullogh (1992, Section 3.5). I don't know the exact formulation you are using, but maybe you could consider an analogue of you model using gls, i.e., lme(..., random=~1|id) gls(..., corr=corCompSymm(form=~1|id)) The references mentioned above are: @Book{verbeke.molenberghs:00, author= {G. Verbeke and G. Molenberghs}, title = {Linear Mixed Models for Longitudinal Data}, year = {2000}, address = {New York}, publisher = {Springer-Verlag} } @Book{searle.et.al:92, author= {S. Searle and G. Cassela and C. McCulloch}, title = {Variance Components}, year = {1992}, address = {New York}, publisher = {Wiley} } I hope it helps. Best, Dimitris Dimitris Rizopoulos Ph.D. Student Biostatistical Centre School of Public Health Catholic University of Leuven Address: Kapucijnenvoer 35, Leuven, Belgium Tel: +32/16/396887 Fax: +32/16/337015 Web: http://www.med.kuleuven.ac.be/biostat/ http://www.student.kuleuven.ac.be/~m0390867/dimitris.htm - Original Message - From: Steve Roberts [EMAIL PROTECTED] To: [EMAIL PROTECTED] Sent: Tuesday, September 21, 2004 1:38 PM Subject: [R] lme RE variance computation As I understand it lme (in R v1.9.x) estimates random effect variances on a log scale, constraining them to be positive. Whilst this seems sensible, it does lead to apparently biased estimates if the variance is actually zero - which makes our simulation results look strange. Whilst we need to think a bit deeper about it - I still haven't got my head around what a negative variance could mean - does anyone know a way to take away the contraint and allowing zero or negative variances? Steve. Dr Steve Roberts [EMAIL PROTECTED] Senior Lecturer in Medical Statistics, CMMCH NHS Trust and University of Manchester Biostatistics Group, 0161 275 5192 / 0161 276 5785 __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Spencer Graves, PhD, Senior Development Engineer O: (408)938-4420; mobile: (408)655-4567 __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Lme warning with coef()
Hello I routinely use the following without a problem: fm1-lme(score~time,data,random=~time|ID) tmp-coef(fm1, augFrame=T) However, in my current situation, I am running into the following error when I execute the coef() call Error in tapply(as.character(object[[nm]]), groups, FUN[[dClass]]) : arguments must have same length There are some cases with missing data, but I believe this was true in prior successful attempts (although I do not recall exactly). Any thoughts on this? Thanks, Harold [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] constrained optimization in R
R: I need to minimize a function such that the parameters when used in another function result in a particular value, which i fix. That is, I need min(f(a,b)) given g(a,b)=X, where min(.) is the minimum, f(.) and g(.) are functions (with unknown gradients) of parameters a and b and X is a fixed value. What optimization function(s) in R do you suggest? constrOptim looks like it will work except I don't know the gradient of the functions. Thanks, Jason Higbee [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] R 1.9.1 Fails to Start on WinXP SP2
On Tue, 21 Sep 2004 09:59:31 -0500, Scott Higginbotham [EMAIL PROTECTED] wrote : I have two computers both of which are running Windows XP SP2. R 1.9.1 runs just fine on one but not the other. Both installations of R came from the same installation package. When I click on the desktop icon an hourglass shows up then disappears. Similarly, if I start RGUI the process shows up in task-manager then abruptly disappears. Interestingly, I can access R in Excel via the D(COM) Server on both computers. Any ideas? Not a one! Could you please try the beta build available on CRAN at http://cran.r-project.org/bin/windows/base/rdevel.html Thanks! Duncan Murdoch __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] constrained optimization in R
Have you considered doing an unconstrained penalized optimization, e.g. pen.f(x, X, penalty) = function(x)(f(x[1], x[2])+penalty*(g(x[1], x[2])-X)^2) Then give pen.f to optim. Start out with penalty small, then use optim's answers with one value of penalty as starting values for optim with double or 10 times the penalty. As long as f and g are reasonably well behaved, this should produce an answer without excessive effort. hope this helps. spencer graves [EMAIL PROTECTED] wrote: R: I need to minimize a function such that the parameters when used in another function result in a particular value, which i fix. That is, I need min(f(a,b)) given g(a,b)=X, where min(.) is the minimum, f(.) and g(.) are functions (with unknown gradients) of parameters a and b and X is a fixed value. What optimization function(s) in R do you suggest? constrOptim looks like it will work except I don't know the gradient of the functions. Thanks, Jason Higbee [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Spencer Graves, PhD, Senior Development Engineer O: (408)938-4420; mobile: (408)655-4567 __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] R 1.9.1 Fails to Start on WinXP SP2
I should have noted in my first post that I had already tried that to no avail. Scott -Original Message- From: Duncan Murdoch [mailto:[EMAIL PROTECTED] Sent: Tuesday, September 21, 2004 1:35 PM To: Scott Higginbotham Cc: [EMAIL PROTECTED] Subject: Re: [R] R 1.9.1 Fails to Start on WinXP SP2 On Tue, 21 Sep 2004 09:59:31 -0500, Scott Higginbotham [EMAIL PROTECTED] wrote : I have two computers both of which are running Windows XP SP2. R 1.9.1 runs just fine on one but not the other. Both installations of R came from the same installation package. When I click on the desktop icon an hourglass shows up then disappears. Similarly, if I start RGUI the process shows up in task-manager then abruptly disappears. Interestingly, I can access R in Excel via the D(COM) Server on both computers. Any ideas? Not a one! Could you please try the beta build available on CRAN at http://cran.r-project.org/bin/windows/base/rdevel.html Thanks! Duncan Murdoch __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] asypow.noncent: how does it work?
[EMAIL PROTECTED] wrote: I am trying to do power calculations for the proportional odds model using the asypow library. The code noncenta90b10-asypow.noncent(theta.ha=a9010,info.mat=infomatrixa90b10,constraints=constrt) returns Error in max(..., na.rm = na.rm) : invalid mode of argument. the various arguments I've used are: a9010 [,1] [1,] -1.7357568 [2,] -0.1928619 specifying the theta.ha array as a row not a column makes no difference infomatrixa90b10 [,1] [,2] [1,] 0.967005807 -0.004699262 [2,] -0.004699262 0.903852346 constrt [,1] [,2] [,3] [1,] 1 a -1.92861865194525 [2,] 1 b 0 I'm probably missing something very simple, but I can't see what it is. Can anyone help? Problem is with your constarint matrix. From the help file (which admittedlt could be clearer) constrain is a 3-column matrix. First element in a row should be 1 to indicate a param set == to a value, 2 to indicate equality of two params. You put 1, so we assume first case. Second row element should be , in this case, index of param set to value, and cannot be a or b. Lets suppose yours are first and second parameter, then it shoud be 1 and 2, respectively. Finally, in this case, third row element should be value. So you should use something like constrt - matrix( c(1,1,-1.9286, 1,2,0), 2, 3, byrow=TRUE) and then, using your values for the other arguments, asypow.noncent(a9010, infomatrix, constrt) $w [,1] [1,] 0.06993048 $df [1] 2 Kjetil -- Kjetil Halvorsen. Peace is the most effective weapon of mass construction. -- Mahdi Elmandjra __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] constrained optimization in R
On Tue, 21 Sep 2004 [EMAIL PROTECTED] wrote: R: I need to minimize a function such that the parameters when used in another function result in a particular value, which i fix. That is, I need min(f(a,b)) given g(a,b)=X, where min(.) is the minimum, f(.) and g(.) are functions (with unknown gradients) of parameters a and b and X is a fixed value. What optimization function(s) in R do you suggest? constrOptim looks like it will work except I don't know the gradient of the functions. constrOptim won't work -- it does INequality constraints. -thomas Thanks, Jason Higbee [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Thomas Lumley Assoc. Professor, Biostatistics [EMAIL PROTECTED] University of Washington, Seattle __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Ever see a stata import problem like this?
Greetings Everybody: I generated a 1.2MB dta file based on the general social survey with Stata8 for linux. The file can be re-opened with Stata, but when I bring it into R, it says all the values are missing for most of the variables. This dataset is called morgen.dta and I dropped a copy online in case you are interested http://www.ku.edu/~pauljohn/R/morgen.dta looks like this to R (I tried various options on the read.dta command): myDat - read.dta(morgen.dta) summary(myDat) CASEID yearid hrs1 hrs2 Min. : 19721 Min. :1972 Min. : 1 NAP :0 NAP :0 1st Qu.: 1983475 1st Qu.:1978 1st Qu.: 445 DK :0 DK :0 Median : 1996808 Median :1987 Median : 905 NA :0 NA :0 Mean : 9963040 Mean :1986 Mean : 990 NA's:40933 NA's:40933 3rd Qu.:19872187 3rd Qu.:1994 3rd Qu.:1358 Max. :20002817 Max. :2000 Max. :3247 prestige agewedage educpaeduc DK,NA,NAP:0 NAP :0 DK :0 NAP :0 NAP :0 NA's :40933 DK :0 NA :0 DK :0 DK :0 NA :0 NA's:40933 NA :0 NA :0 NA's:40933NA's:40933 NA's:40933 maeduc speduc income NAP :0 NAP :0 $25000 OR MORE:14525 DK :0 DK :0 $1 - 14999: 5022 NA :0 NA :0 $15000 - 1: 3869 NA's:40933 NA's:40933 $2 - 24999: 3664 REFUSED : 1877 (Other) : 8523 NA's : 3453 Here's what Stata sees when I load the same thing: summarize, detail Case identification number - Percentiles Smallest 1% 197432 19721 5% 199649 19722 10% 1974116 19723 Obs 40933 25% 1983475 19724 Sum of Wgt. 40933 50% 1996808 Mean9963040 Largest Std. Dev. 9006352 75% 1.99e+07 2.00e+07 90% 2.00e+07 2.00e+07 Variance 8.11e+13 95% 2.00e+07 2.00e+07 Skewness .18931 99% 2.00e+07 2.00e+07 Kurtosis 1.045409 GSS YEAR FOR THIS RESPONDENT - Percentiles Smallest 1% 1972 1972 5% 1973 1972 10% 1974 1972 Obs 40933 25% 1978 1972 Sum of Wgt. 40933 50% 1987 Mean 1986.421 Largest Std. Dev. 8.61136 75% 1994 2000 90% 1998 2000 Variance 74.15552 95% 2000 2000 Skewness -.0789223 99% 2000 2000 Kurtosis 1.799939 RESPONDENT ID NUMBER - Percentiles Smallest 1% 18 1 5% 89 1 10% 178 1 Obs 40933 25% 445 1 Sum of Wgt. 40933 50% 905 Mean 989.9129 Largest Std. Dev. 689.0596 75% 1358 3244 90% 2027 3245 Variance 474803.2 95% 2437 3246 Skewness .8359211 99% 2867 3247 Kurtosis 3.311248 NUMBER OF HOURS WORKED LAST WEEK - Percentiles Smallest 1%6 0 5% 15 0 10% 21 0 Obs 23279 25% 37 0 Sum of Wgt. 23279 50% 40 Mean 41.05206 Largest Std. Dev. 13.95931 75% 48 89 90% 60 89 Variance 194.8624 95% 65 89 Skewness.195045 99% 82 89 Kurtosis 4.448998 NUMBER OF HOURS USUALLY WORK A WEEK - Percentiles Smallest 1%4 0 5% 15 0 10% 20 1 Obs 774 25% 38 2 Sum of Wgt. 774 50% 40 Mean 39.79199 Largest Std. Dev. 13.43383 75% 45 89 90% 55 89 Variance 180.4677 95% 60 89
[R] RMySQL and Blob
Dear R experts, Does RMySQL package handle Blob datatype in a MySQL database? Blob can represent an image, a sound or some other large and complex binary objects. In an article published by R-database special interest group, named A common database interface (DBI) (updated June 2003), it's mentioned in open issues and limitations that We need to carefully plan how to deal with binary objects. Before I invest time to try, I would appreciate any experts' opinions. Thanks, Jonathan __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Ever see a stata import problem like this?
On Tue, 21 Sep 2004, Paul Johnson wrote: Greetings Everybody: I generated a 1.2MB dta file based on the general social survey with Stata8 for linux. The file can be re-opened with Stata, but when I bring it into R, it says all the values are missing for most of the variables. You need read.dta( ,convert.factors=FALSE) You have variables with labels for some, but not all, of their values. When these are converted to R factors you lose the unlabelled values. R does not have a data type that is sometimes labelled and sometimes numeric. When you use convert.factors=FALSE the label information is still read in and returned as an attribute of the data frame, so you can set individual variables to be factors. -thomas This dataset is called morgen.dta and I dropped a copy online in case you are interested http://www.ku.edu/~pauljohn/R/morgen.dta looks like this to R (I tried various options on the read.dta command): myDat - read.dta(morgen.dta) summary(myDat) CASEID yearid hrs1 hrs2 Min. : 19721 Min. :1972 Min. : 1 NAP :0 NAP :0 1st Qu.: 1983475 1st Qu.:1978 1st Qu.: 445 DK :0 DK :0 Median : 1996808 Median :1987 Median : 905 NA :0 NA :0 Mean : 9963040 Mean :1986 Mean : 990 NA's:40933 NA's:40933 3rd Qu.:19872187 3rd Qu.:1994 3rd Qu.:1358 Max. :20002817 Max. :2000 Max. :3247 prestige agewedage educpaeduc DK,NA,NAP:0 NAP :0 DK :0 NAP :0 NAP :0 NA's :40933 DK :0 NA :0 DK :0 DK :0 NA :0 NA's:40933 NA :0 NA :0 NA's:40933NA's:40933 NA's:40933 maeduc speduc income NAP :0 NAP :0 $25000 OR MORE:14525 DK :0 DK :0 $1 - 14999: 5022 NA :0 NA :0 $15000 - 1: 3869 NA's:40933 NA's:40933 $2 - 24999: 3664 REFUSED : 1877 (Other) : 8523 NA's : 3453 Here's what Stata sees when I load the same thing: summarize, detail Case identification number - Percentiles Smallest 1% 197432 19721 5% 199649 19722 10% 1974116 19723 Obs 40933 25% 1983475 19724 Sum of Wgt. 40933 50% 1996808 Mean9963040 Largest Std. Dev. 9006352 75% 1.99e+07 2.00e+07 90% 2.00e+07 2.00e+07 Variance 8.11e+13 95% 2.00e+07 2.00e+07 Skewness .18931 99% 2.00e+07 2.00e+07 Kurtosis 1.045409 GSS YEAR FOR THIS RESPONDENT - Percentiles Smallest 1% 1972 1972 5% 1973 1972 10% 1974 1972 Obs 40933 25% 1978 1972 Sum of Wgt. 40933 50% 1987 Mean 1986.421 Largest Std. Dev. 8.61136 75% 1994 2000 90% 1998 2000 Variance 74.15552 95% 2000 2000 Skewness -.0789223 99% 2000 2000 Kurtosis 1.799939 RESPONDENT ID NUMBER - Percentiles Smallest 1% 18 1 5% 89 1 10% 178 1 Obs 40933 25% 445 1 Sum of Wgt. 40933 50% 905 Mean 989.9129 Largest Std. Dev. 689.0596 75% 1358 3244 90% 2027 3245 Variance 474803.2 95% 2437 3246 Skewness .8359211 99% 2867 3247 Kurtosis 3.311248 NUMBER OF HOURS WORKED LAST WEEK - Percentiles Smallest 1%6 0 5% 15 0 10% 21 0 Obs 23279 25% 37 0 Sum of Wgt. 23279 50% 40 Mean 41.05206 Largest Std. Dev. 13.95931 75% 48 89 90% 60 89 Variance 194.8624 95% 65 89 Skewness.195045 99% 82 89 Kurtosis 4.448998 NUMBER OF HOURS USUALLY WORK A WEEK -
Re: [R] R 1.9.1 Fails to Start on WinXP SP2
On Tue, 21 Sep 2004 14:15:51 -0500, Scott Higginbotham [EMAIL PROTECTED] wrote: I should have noted in my first post that I had already tried that to no avail. In that case, things aren't easy. You need to see where in the startup sequence it's bailing out. We don't have debug code installed to do that (because this isn't a common problem!), so the only way I know to find that is to compile R with debug information, and run it under a debugger. If you've never done that, it's a big job. Duncan Murdoch Scott -Original Message- From: Duncan Murdoch [mailto:[EMAIL PROTECTED] Sent: Tuesday, September 21, 2004 1:35 PM To: Scott Higginbotham Cc: [EMAIL PROTECTED] Subject: Re: [R] R 1.9.1 Fails to Start on WinXP SP2 On Tue, 21 Sep 2004 09:59:31 -0500, Scott Higginbotham [EMAIL PROTECTED] wrote : I have two computers both of which are running Windows XP SP2. R 1.9.1 runs just fine on one but not the other. Both installations of R came from the same installation package. When I click on the desktop icon an hourglass shows up then disappears. Similarly, if I start RGUI the process shows up in task-manager then abruptly disappears. Interestingly, I can access R in Excel via the D(COM) Server on both computers. Any ideas? Not a one! Could you please try the beta build available on CRAN at http://cran.r-project.org/bin/windows/base/rdevel.html Thanks! Duncan Murdoch __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] R 1.9.1 Fails to Start on WinXP SP2
One other suggestion: Run the msconfig System configuration utility to turn off most of the software that loads on your machine at startup, and see if that allows R to start. Then gradually add it back until you find the culprit, if there is one. Duncan Murdoch __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] R 1.9.1 Fails to Start on WinXP SP2
Perhaps a couple of things to try before that: 1. See what Rterm.exe does. 2. See whether it makes any difference if the --vanilla flag is added to the shortcut to Rgui.exe. (E.g., was Rgui trying to load an existing .rda file or execute some startup code on the computer that had the problem, and otherwise on the other?) Andy From: Duncan Murdoch On Tue, 21 Sep 2004 14:15:51 -0500, Scott Higginbotham [EMAIL PROTECTED] wrote: I should have noted in my first post that I had already tried that to no avail. In that case, things aren't easy. You need to see where in the startup sequence it's bailing out. We don't have debug code installed to do that (because this isn't a common problem!), so the only way I know to find that is to compile R with debug information, and run it under a debugger. If you've never done that, it's a big job. Duncan Murdoch Scott -Original Message- From: Duncan Murdoch [mailto:[EMAIL PROTECTED] Sent: Tuesday, September 21, 2004 1:35 PM To: Scott Higginbotham Cc: [EMAIL PROTECTED] Subject: Re: [R] R 1.9.1 Fails to Start on WinXP SP2 On Tue, 21 Sep 2004 09:59:31 -0500, Scott Higginbotham [EMAIL PROTECTED] wrote : I have two computers both of which are running Windows XP SP2. R 1.9.1 runs just fine on one but not the other. Both installations of R came from the same installation package. When I click on the desktop icon an hourglass shows up then disappears. Similarly, if I start RGUI the process shows up in task-manager then abruptly disappears. Interestingly, I can access R in Excel via the D(COM) Server on both computers. Any ideas? Not a one! Could you please try the beta build available on CRAN at http://cran.r-project.org/bin/windows/base/rdevel.html Thanks! Duncan Murdoch __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] ordered probit and cauchit
What is the current state of the R-art for ordered probit models, and more esoterically is there any available R strategy for ordered cauchit models, i.e. ordered multinomial alternatives with a cauchy link function. MCMC is an option, obviously, but for a univariate latent variable model this seems to be overkill... standard mle methods should be preferable. (??) Googling reveals that spss provides such functions... just to wave a red flag. url:www.econ.uiuc.edu/~rogerRoger Koenker email [EMAIL PROTECTED] Department of Economics vox:217-333-4558University of Illinois fax:217-244-6678Champaign, IL 61820 __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] png problem
Hi Clément Calenge wrote: Dear R-users, I have a small problem with the function png(), when used with the argument colortype=pseudo.cube. png(toto.png, colortype=pseudo.cube) image(matrix(rnorm(1), 100, 100)) dev.off() R is blocked at the last command (R does not print any prompt after the last command). Nothing is written in the file (Gimp indicates that the file is corrupted). Did you wait long enough? This example took a little while to complete for me (may need someone more familiar with the code to tell us why it is so slow). Paul However, png(toto.png) image(matrix(rnorm(1), 100, 100)) dev.off() works fine. I tried: options(X11colortype = pseudo.cube) png(toto.png) image(matrix(rnorm(1), 100, 100)) dev.off() But, here again, R is blocked. I tried to replace dev.off() by graphics.off(), but this does not resolve the problem. The problem does not occurs when the function X11() is used instead of the function png(). I searched through the mail archive, the FAQ, on google, but I did not found any solution to this problem. On the help page on the function png(), it is indicated that The colour handling will be that of the 'X11' device in use. I never used these functions before, but maybe png() is not suitable with colortype=pseudo.cube ? Can you tell me where I have missed something ? Thanks in Advance, Clément Calenge. version _ platform sparc-sun-solaris2.9 arch sparc os solaris2.9 system sparc, solaris2.9 status major1 minor9.1 year 2004 month06 day 21 language R [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Dr Paul Murrell Department of Statistics The University of Auckland Private Bag 92019 Auckland New Zealand 64 9 3737599 x85392 [EMAIL PROTECTED] http://www.stat.auckland.ac.nz/~paul/ __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] lda predict
On Tue, 21 Sep 2004, rob foxall (IFR) wrote: Dear R-helpers, I have a model created by lda, and I would like to use this model to make predictions for new or old data. The catch is, I want to do this without using the predict function, i.e. only using information directly from the foo.lda object to create my posterior probabilities. Well, that's what predict.lda does, so please read its code. In anticipation of likely responses, I will be brushing up my lda knowledge using the given references when I have time, but am being hassled for an answer asap! -- 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 __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html