[R] Assistance with data import from Statistica
I am a new R user attempting to convert files from Statistica to R. I can export from Statistica to SPSS .por format, but not to the SPSS .sav format. Is there a procedure for easily accomplishing this, which will allow me to keep variable short and long labels (big surveys LOTS of time to replace all this work). Many thanks for sharing your time and knowledge. steve Dr Steven M Burgess Associate Director (Research) Professor of Business Administration in Marketing Graduate School of Business University of Cape Town Breakwater Campus, Portswood Road Greenpoint, Cape Town 8001 South Africa Office: Tel +27-21-406-1416, fax +27-21-406-1412 Mobile: Tel +27-83-226-2811 Home:Tel +27-21-791-3423, fax +27-21-791-3425 Confidentiality: This e-mail message is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorised review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. I do not represent, warrant and/or guarantee that the integrity of this communication has been maintained nor that the communication is free of virus, interception or interference or errors caused by virus, interception or interference. [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Assistance with data import from Statistica
I'm never work with Stats but IMHO you find in library(foreign) all what you need. cheers,christian Am Sonntag, 1. Februar 2004 09:39 schrieb Steve Burgess: I am a new R user attempting to convert files from Statistica to R. I can export from Statistica to SPSS .por format, but not to the SPSS .sav format. Is there a procedure for easily accomplishing this, which will allow me to keep variable short and long labels (big surveys LOTS of time to replace all this work). Many thanks for sharing your time and knowledge. steve Dr Steven M Burgess Associate Director (Research) Professor of Business Administration in Marketing Graduate School of Business University of Cape Town Breakwater Campus, Portswood Road Greenpoint, Cape Town 8001 South Africa Office: Tel +27-21-406-1416, fax +27-21-406-1412 Mobile: Tel +27-83-226-2811 Home:Tel +27-21-791-3423, fax +27-21-791-3425 Confidentiality: This e-mail message is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorised review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. I do not represent, warrant and/or guarantee that the integrity of this communication has been maintained nor that the communication is free of virus, interception or interference or errors caused by virus, interception or interference. [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.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://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] I can't make .C(...) web-page example.
Hi everyone! I'm trying to lear how to call external C code in R but even the R help web-page example is drive me crazy. I copy-paste the example at: http://stat.ethz.ch/R-manual/R-patched/doc/manual/R-exts.html#Interface%20functions%20.C%20and%20.Fortran here is the example: void convolve(double *a, int *na, double *b, int *nb, double *ab) { int i, j, nab = *na + *nb - 1; for(i = 0; i nab; i++) ab[i] = 0.0; for(i = 0; i *na; i++) for(j = 0; j *nb; j++) ab[i + j] += a[i] * b[j]; } called from R by conv - function(a, b) .C(convolve, as.double(a), as.integer(length(a)), as.double(b), as.integer(length(b)), ab = double(length(a) + length(b) - 1))$ab but using the conv function in simple examples give me trouble: conv(1:10,2:11) Error in conv(1:10, 2:11) : pairlist object cannot be coerced to double conv( rnorm(10,10,1) , rnorm(10,11,1) ) Segmentation fault conv( c(10.53, 8.456, 6.6) , c(10.53, 8.456, 6.6) ) Error in conv(c(10.53, 8.456, 6.6), c(10.53, 8.456, 6.6)) : cannot allocate vector of length 1717986918 a = c(1.46756, 5.56456, 2.3646) b = a + 5.4 conv(a,b) Error in conv(a, b) : cannot coerce type promise to double vector and so on. These results appear if I have only the .c file or if I use gcc -c conv.c to create .o file. I cannot create executable because there is no main(...) in .c file but just the copy-pasted example. Someone knows what is happening to me ? __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] coxph in R
Where are the estimates of the baseline hazard in the output of coxph? Thanks, Jim James S. Clark H.L. Blomquist Professor Department of Biology and Nicholas School of the Environment Duke University Durham, NC 27708 Biology: http://www.biology.duke.edu/research_by_area/eeob/clark.html Center on Global Change: http://www.env.duke.edu/cgc/ Ecology Program: http://www.ecology.duke.edu/ __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] I can't make .C(...) web-page example.
Change the name. There is a convolve C function in R 1.8.x (but not R-devel) that is being found rather than yours. On Sun, 1 Feb 2004, Ricardo Zorzetto Nicoliello Vencio wrote: Hi everyone! I'm trying to lear how to call external C code in R but even the R help web-page example is drive me crazy. I copy-paste the example at: http://stat.ethz.ch/R-manual/R-patched/doc/manual/R-exts.html#Interface%20functions%20.C%20and%20.Fortran here is the example: void convolve(double *a, int *na, double *b, int *nb, double *ab) { int i, j, nab = *na + *nb - 1; for(i = 0; i nab; i++) ab[i] = 0.0; for(i = 0; i *na; i++) for(j = 0; j *nb; j++) ab[i + j] += a[i] * b[j]; } called from R by conv - function(a, b) .C(convolve, as.double(a), as.integer(length(a)), as.double(b), as.integer(length(b)), ab = double(length(a) + length(b) - 1))$ab but using the conv function in simple examples give me trouble: conv(1:10,2:11) Error in conv(1:10, 2:11) : pairlist object cannot be coerced to double conv( rnorm(10,10,1) , rnorm(10,11,1) ) Segmentation fault conv( c(10.53, 8.456, 6.6) , c(10.53, 8.456, 6.6) ) Error in conv(c(10.53, 8.456, 6.6), c(10.53, 8.456, 6.6)) : cannot allocate vector of length 1717986918 a = c(1.46756, 5.56456, 2.3646) b = a + 5.4 conv(a,b) Error in conv(a, b) : cannot coerce type promise to double vector and so on. These results appear if I have only the .c file or if I use gcc -c conv.c to create .o file. I cannot create executable because there is no main(...) in .c file but just the copy-pasted example. Someone knows what is happening to me ? __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- 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://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] coxph
Where are the estimates of the baseline hazard for coxph? Thanks, Jim James S. Clark H.L. Blomquist Professor Department of Biology and Nicholas School of the Environment Duke University Durham, NC 27708 Biology: http://www.biology.duke.edu/research_by_area/eeob/clark.html Center on Global Change: http://www.env.duke.edu/cgc/ Ecology Program: http://www.ecology.duke.edu/ __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] I can't make .C(...) web-page example.
Hi! I am by no means an R-expert nor a C-expert but the document An Introduction to .C Interface to R by Roger D. Peng and Jan de Leeuw helped me a lot to find out how it works in principle to use the .C function call. You can find it on the homepage of Roger D. Peng (http://www.biostat.jhsph.edu/~rpeng/ ) at the following URL: http://www.biostat.jhsph.edu/~rpeng/docs/interface.pdf Hope this is what you were looking for. Best, Roland -Original Message- From: Ricardo Zorzetto Nicoliello Vencio [SMTP:[EMAIL PROTECTED] Sent: Sunday, February 01, 2004 3:56 PM To: [EMAIL PROTECTED] Subject: [R] I can't make .C(...) web-page example. Hi everyone! I'm trying to lear how to call external C code in R but even the R help web-page example is drive me crazy. I copy-paste the example at: http://stat.ethz.ch/R-manual/R-patched/doc/manual/R-exts.html#Interface%20 functions%20.C%20and%20.Fortran here is the example: void convolve(double *a, int *na, double *b, int *nb, double *ab) { int i, j, nab = *na + *nb - 1; for(i = 0; i nab; i++) ab[i] = 0.0; for(i = 0; i *na; i++) for(j = 0; j *nb; j++) ab[i + j] += a[i] * b[j]; } called from R by conv - function(a, b) .C(convolve, as.double(a), as.integer(length(a)), as.double(b), as.integer(length(b)), ab = double(length(a) + length(b) - 1))$ab but using the conv function in simple examples give me trouble: conv(1:10,2:11) Error in conv(1:10, 2:11) : pairlist object cannot be coerced to double conv( rnorm(10,10,1) , rnorm(10,11,1) ) Segmentation fault conv( c(10.53, 8.456, 6.6) , c(10.53, 8.456, 6.6) ) Error in conv(c(10.53, 8.456, 6.6), c(10.53, 8.456, 6.6)) : cannot allocate vector of length 1717986918 a = c(1.46756, 5.56456, 2.3646) b = a + 5.4 conv(a,b) Error in conv(a, b) : cannot coerce type promise to double vector and so on. These results appear if I have only the .c file or if I use gcc -c conv.c to create .o file. I cannot create executable because there is no main(...) in .c file but just the copy-pasted example. Someone knows what is happening to me ? __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html + This mail has been sent through the MPI for Demographic Research. Should you receive a mail that is apparently from a MPI user without this text displayed, then the address has most likely been faked. If you are uncertain about the validity of this message, please check the mail header or ask your system administrator for assistance. __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] I can't make .C(...) web-page example.
Ricardo Zorzetto Nicoliello Vencio wrote: Hi everyone! I'm trying to lear how to call external C code in R but even the R help web-page example is drive me crazy. I copy-paste the example at: http://stat.ethz.ch/R-manual/R-patched/doc/manual/R-exts.html#Interface%20functions%20.C%20and%20.Fortran here is the example: void convolve(double *a, int *na, double *b, int *nb, double *ab) { int i, j, nab = *na + *nb - 1; for(i = 0; i nab; i++) ab[i] = 0.0; for(i = 0; i *na; i++) for(j = 0; j *nb; j++) ab[i + j] += a[i] * b[j]; } called from R by conv - function(a, b) .C(convolve, as.double(a), as.integer(length(a)), as.double(b), as.integer(length(b)), ab = double(length(a) + length(b) - 1))$ab but using the conv function in simple examples give me trouble: conv(1:10,2:11) Error in conv(1:10, 2:11) : pairlist object cannot be coerced to double conv( rnorm(10,10,1) , rnorm(10,11,1) ) Segmentation fault conv( c(10.53, 8.456, 6.6) , c(10.53, 8.456, 6.6) ) Error in conv(c(10.53, 8.456, 6.6), c(10.53, 8.456, 6.6)) : cannot allocate vector of length 1717986918 a = c(1.46756, 5.56456, 2.3646) b = a + 5.4 conv(a,b) Error in conv(a, b) : cannot coerce type promise to double vector and so on. These results appear if I have only the .c file or if I use gcc -c conv.c to create .o file. I cannot create executable because there is no main(...) in .c file but just the copy-pasted example. Most easily, use the R wrapper to compile and link files as in: R CMD SHLIB convolve.c which produces an .so file you can dyn.load(). (If you are on Windows: Rcmd SHLIB makes a dll.) Uwe Ligges Someone knows what is happening to me ? __ [EMAIL PROTECTED] mailing list https://www.stat.math.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://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] CART: rapart vs bagging
Hi, Is here anyone knows the difference between rapart and bagging when grow a CART tree? Thanks Qin __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] 3 little questions
Dear R-helpers, 3 questions: 1. Is there a package that contains a routine for computing Kendall's W (coefficient of concordance), with and without ties ? 2. Is there a package that contains a routine for computing Goodman' s Gamma. 3. I there a simple method for computing the number of ties as well as their lengths within a vector fo ranks, e.g. r1 - rank(c(1, 3, 2, 3, 3, 2, 4)) gives: [1] 1.0 5.0 2.5 5.0 5.0 2.5 7.0 which contains 2 ties with length 2 and 3. Thanks in advance, S.M == Dr. Siegfried Macho Department of Psychology University of Fribourg Rue de Faucigny 2 CH-1700 Fribourg Tel.: +41-26-3007635 Fax.: +41-26-3009712 http://www.unifr.ch/psycho/general/macho.htm __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] I can't make .C(...) web-page example.
Ricardo Zorzetto Nicoliello Vencio [EMAIL PROTECTED] writes: Hi everyone! I'm trying to lear how to call external C code in R but even the R help web-page example is drive me crazy. I copy-paste the example at: http://stat.ethz.ch/R-manual/R-patched/doc/manual/R-exts.html#Interface%20functions%20.C%20and%20.Fortran here is the example: void convolve(double *a, int *na, double *b, int *nb, double *ab) { int i, j, nab = *na + *nb - 1; for(i = 0; i nab; i++) ab[i] = 0.0; for(i = 0; i *na; i++) for(j = 0; j *nb; j++) ab[i + j] += a[i] * b[j]; } called from R by conv - function(a, b) .C(convolve, as.double(a), as.integer(length(a)), as.double(b), as.integer(length(b)), ab = double(length(a) + length(b) - 1))$ab but using the conv function in simple examples give me trouble: conv(1:10,2:11) Error in conv(1:10, 2:11) : pairlist object cannot be coerced to double conv( rnorm(10,10,1) , rnorm(10,11,1) ) Segmentation fault conv( c(10.53, 8.456, 6.6) , c(10.53, 8.456, 6.6) ) Error in conv(c(10.53, 8.456, 6.6), c(10.53, 8.456, 6.6)) : cannot allocate vector of length 1717986918 a = c(1.46756, 5.56456, 2.3646) b = a + 5.4 conv(a,b) Error in conv(a, b) : cannot coerce type promise to double vector and so on. These results appear if I have only the .c file or if I use gcc -c conv.c to create .o file. I cannot create executable because there is no main(...) in .c file but just the copy-pasted example. Someone knows what is happening to me ? Looks like you haven't quite understood how to generate a shared library (R CMD SHLIB) and load it dynamically (dyn.load), and then somehow pick up a different convolve entry point (doesn't seem possible with Linux, but you're not tellig us what your system is...). Either that or you managed to confuse R badly through some earlier attempts to load the module. -- 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://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] coxph in R
Nowhere. They are not in Cox's (1972) paper either! (One place is in the discussion of that paper, but the point is that they are an orthogonal concept to partial likelihood estimation of the coefficients.) You use survfit() on a coxph fit object to compute them. On Sun, 1 Feb 2004, Jim Clark wrote: Where are the estimates of the baseline hazard in the output of coxph? Thanks, Jim James S. Clark H.L. Blomquist Professor Department of Biology and Nicholas School of the Environment Duke University Durham, NC 27708 Biology: http://www.biology.duke.edu/research_by_area/eeob/clark.html Center on Global Change: http://www.env.duke.edu/cgc/ Ecology Program: http://www.ecology.duke.edu/ -- 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://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] coxph
Jim Clark [EMAIL PROTECTED] writes: [twice...] Where are the estimates of the baseline hazard for coxph? That's not an estimable quantity. However, estimates of the integrated hazard or the survival function can be obtained with basehaz(fit) resp. survfit(fit). -- 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://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] I can't make .C(...) web-page example.
On 1 Feb 2004, Peter Dalgaard wrote: Ricardo Zorzetto Nicoliello Vencio [EMAIL PROTECTED] writes: Hi everyone! I'm trying to lear how to call external C code in R but even the R help web-page example is drive me crazy. I copy-paste the example at: http://stat.ethz.ch/R-manual/R-patched/doc/manual/R-exts.html#Interface%20functions%20.C%20and%20.Fortran here is the example: void convolve(double *a, int *na, double *b, int *nb, double *ab) { int i, j, nab = *na + *nb - 1; for(i = 0; i nab; i++) ab[i] = 0.0; for(i = 0; i *na; i++) for(j = 0; j *nb; j++) ab[i + j] += a[i] * b[j]; } called from R by conv - function(a, b) .C(convolve, as.double(a), as.integer(length(a)), as.double(b), as.integer(length(b)), ab = double(length(a) + length(b) - 1))$ab but using the conv function in simple examples give me trouble: conv(1:10,2:11) Error in conv(1:10, 2:11) : pairlist object cannot be coerced to double conv( rnorm(10,10,1) , rnorm(10,11,1) ) Segmentation fault conv( c(10.53, 8.456, 6.6) , c(10.53, 8.456, 6.6) ) Error in conv(c(10.53, 8.456, 6.6), c(10.53, 8.456, 6.6)) : cannot allocate vector of length 1717986918 a = c(1.46756, 5.56456, 2.3646) b = a + 5.4 conv(a,b) Error in conv(a, b) : cannot coerce type promise to double vector and so on. These results appear if I have only the .c file or if I use gcc -c conv.c to create .o file. I cannot create executable because there is no main(...) in .c file but just the copy-pasted example. Someone knows what is happening to me ? Looks like you haven't quite understood how to generate a shared library (R CMD SHLIB) and load it dynamically (dyn.load), and then somehow pick up a different convolve entry point (doesn't seem possible with Linux, but you're not tellig us what your system is...). Either that or you managed to confuse R badly through some earlier attempts to load the module. There is an entry point convolve in package ts which is nowadays loaded by default. A week or so ago that one was being found under Windows even after dynamically loading convolve.dll, so I renamed it in R-devel. I got the first error message listed above with that example. If you try another name (convolve1, for example) it should either work or report that the symbol cannot be found, hence my suggestion to use another name. -- 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://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] coxph
On 1 Feb 2004, Peter Dalgaard wrote: Jim Clark [EMAIL PROTECTED] writes: [twice...] Where are the estimates of the baseline hazard for coxph? That's not an estimable quantity. However, estimates of the integrated hazard or the survival function can be obtained with basehaz(fit) resp. survfit(fit). Not estimable? Well, neither is the cumulative hazard/survival fnction then, as you only get estimates at the event times. In both cases you need further assumptions on the hazard, which can be smoothness or sum of delta functions or -- 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://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] CART: rapart vs bagging
On Sun, 1 Feb 2004, Qin Liu wrote: Is here anyone knows the difference between rapart and bagging when grow a CART tree? Short answer: `Yes'. Slightly longer answer: `rapart classification tree' is not something Google knows about, and bagging is not to do with `grow a tree'. More informative answer: Bagging is a method to combine ensembles of classification trees, and *rpart* is a package in R which does grow and prune classification trees. CART is the trademark of a commercial program to grow and prune classification trees. Put `bagging Breiman' into Google to help you with the concept of bagging, and then look at package ipred. -- 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://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] interactive 2-D plot interrogation
Greetings Friends: I am trying to find a widget or some other method to let me interrogate a 2-D image using the mouse. I have a data.frame that I can visualize using image() in R. I would like to point to a pixel and have its coordinates displayed. I've read Peter Dalgaard's primer on R-Tcl/Tk, R-News, 1(3):27-31, and his update R-News2(3):25-27, but am still struggling. Can anyone suggest a method for interactively interrogating of a 2-D image? Many thanks. Charles Annis, P.E. [EMAIL PROTECTED] phone: 561-352-9699 eFax: 503-217-4849 http://www.StatisticalEngineering.com __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Stepwise Regression and PLS
Dear all, I am a newcomer to R. I intend to using R to do stepwise regression and PLS with a data set (a 55x20 matrix, with one dependent and 19 independent variable). Based on the same data set, I have done the same work using SPSS and SAS. However, there is much difference between the results obtained by R and SPSS or SAS. In the case of stepwise, SPSS gave out a model with 4 independent variable, but with step(), R gave out a model with 10 and much higher R2. Furthermore, regsubsets() also indicate the 10 variable is one of the best regression subset. How to explain this difference? And in the case of my data set, how many variables that enter the model would be reasonable? In the case of PLS, the results of mvr function of pls.pcr package is also different with that of SAS. Although the number of optimum latent variables is same, the difference between R2 is much large. Why? Any comment and suggestion is very appreciated. Thanks in advance! Best wishes, Jinsong Zhao = (Mr.) Jinsong Zhao Ph.D. Candidate School of the Environment Nanjing University 22 Hankou Road, Nanjing 210093 P.R. China E-mail: [EMAIL PROTECTED] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Stepwise Regression and PLS
On Sun, 1 Feb 2004 11:09:28 -0800 (PST) Jinsong Zhao [EMAIL PROTECTED] wrote: Dear all, I am a newcomer to R. I intend to using R to do stepwise regression and PLS with a data set (a 55x20 matrix, with one dependent and 19 independent variable). Based on the same data set, I have done the same work using SPSS and SAS. However, there is much difference between the results obtained by R and SPSS or SAS. In the case of stepwise, SPSS gave out a model with 4 independent variable, but with step(), R gave out a model with 10 and much higher R2. Furthermore, regsubsets() also indicate the 10 variable is one of the best regression subset. How to explain this difference? And in the case of my data set, how many variables that enter the model would be reasonable? In the case of PLS, the results of mvr function of pls.pcr package is also different with that of SAS. Although the number of optimum latent variables is same, the difference between R2 is much large. Why? Any comment and suggestion is very appreciated. Thanks in advance! Best wishes, Jinsong Zhao In your case SPSS, SAS, R, S-Plus, Stata, Systat, Statistica, and every other package will agree in one sense, because results from all of them will be virtually meaningless. Simulate some data from a known model and you'll quickly find out why stepwise variable selection is often a train wreck. --- Frank E Harrell Jr Professor and Chair School of Medicine Department of Biostatistics Vanderbilt University __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] coxph
Prof Brian Ripley [EMAIL PROTECTED] writes: On 1 Feb 2004, Peter Dalgaard wrote: Jim Clark [EMAIL PROTECTED] writes: [twice...] Where are the estimates of the baseline hazard for coxph? That's not an estimable quantity. However, estimates of the integrated hazard or the survival function can be obtained with basehaz(fit) resp. survfit(fit). Not estimable? Well, neither is the cumulative hazard/survival fnction then, as you only get estimates at the event times. In both cases you need further assumptions on the hazard, which can be smoothness or sum of delta functions or Well This *is* quibbling you know. The relation is basically the same as with densities and distribution functions. You can of course, if you want to, define the estimated hazard as a sum of delta functions. However, it won't converge to the true hazard function in any of the standard senses as n increases (although it will in the distribution sense, but that is basically the point of saying that its indefinite integral is estimable). In contrast, you can define the integrated hazard function by extending the value at event times as a right-continuous step function and it will converge pointwise under relatively mild conditions (the censoring mechanism cannot be too harsh and the regressors should behave sensibly). -- 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://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] lines in 3d-cloud plot (lattice)
Hi Staying with Trellis graphics, and depending on what sort of lines you want to add, you could do something similar to persp() example # 2 using cloud() by writing your own panel.3d.cloud() function, which calls panel.3dscatter() and then does some further drawing. This is an analogue to the sequence plot(), followed by points() or lines() in the base plotting functions, all done within a lattice panel function. From a quick look at help(panel.cloud) it looks like the required transformation information is available via something like ltransform3dto3d(). Paul Tom Blackwell wrote: Pascal - Getting away from Trellis graphics, you might consider using persp() with regular graphics. Look at example # 2 under help(persp). - tom blackwell - u michigan medical school - ann arbor - On Fri, 30 Jan 2004, Deepayan Sarkar wrote: On Friday 30 January 2004 06:13, Pascal A. Niklaus wrote: Hi all, I'd like to plot a set of data (x,y,z) as 3D-cloud, and add several line plots to the same 3D graph: Two questions: 1) How do I connect points to get a line? cloud(z~x*y,data=d,zlim=c(0,1))# works cloud(z~x*predict(l),data=d,zlim=c(0,1),type=l) # type=l doesn't Warning message: type = l not implemented, consider using 'panel.3d.cloud = panel.3dscatter.old' in: panel.3d.cloud(x = x, y = y, z = z, rot.mat = rot.mat, za = za, Well, have you considered taking the hint and try cloud(z~x*predict(l),data=d,zlim=c(0,1),type=l, panel.3d.cloud = panel.3dscatter.old) ? help.search(panel.3d.cloud) also didn't report any hits. panel.3d.cloud is the name of an argument to the panel.cloud function. See ?panel.cloud for details. (Unfortunately, the docs are a bit outdated). Briefly, panel.3dscatter.old is a very simple function, that calculates the 2D projections of the given 3D points and then calls panel.xyplot with those. Any 'type' argument which works with panel.xyplot would also work here, including 'p' and 'l'. But no consideration is made of the fact that these are 3D data. For instance, type = 'h' would not give you what you would expect. panel.3dscatter (the newer version) is a bit more sophisticated. For type = 'p', it draws the points in order of increasing depth, so that closer points overwrite distant ones. Unfortunately, a collection of line segments is not well ordered, and I haven't decided yet what to do in that case (which is why the older version is still retained). 2) How do I superimpose a second data set onto the same graph? (something equivalent to the sequence plot(), followed by points() or lines() in the base plotting functions) I'm not sure what you mean. Trellis plots are not supposed to be used for two unrelated data sets, they are typically very much dependent on the structure of the data set. Maybe we could help if you give more details of what exactly you want to do, but before that you should read the ?panel.cloud help page carefully, since anything 'special' would almost invariably involve playing with things documented there. Hth, Deepayan __ [EMAIL PROTECTED] mailing list https://www.stat.math.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://www.stat.math.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://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] interactive 2-D plot interrogation
Hi Depending on how you want the coordinates displayed, the locator() function may help you, or perhaps a simple function written around locator() to process its result into something more meaningful for your siutation. Paul Dirk Eddelbuettel wrote: On Sun, Feb 01, 2004 at 01:53:14PM -0500, Charles Annis, P.E. wrote: Greetings Friends: I am trying to find a widget or some other method to let me interrogate a 2-D image using the mouse. I have a data.frame that I can visualize using image() in R. I would like to point to a pixel and have its coordinates displayed. I've read Peter Dalgaard's primer on R-Tcl/Tk, R-News, 1(3):27-31, and his update R-News2(3):25-27, but am still struggling. Can anyone suggest a method for interactively interrogating of a 2-D image? Look at the InteractiveTkrPlot example on James Wettenhall's site: http://bioinf.wehi.edu.au/~wettenhall/RTclTkExamples/InteractiveTkrPlot.html His example is over discrete set of points, which you can adapt to find the location with respect to curves or lines (which I did at work, where I can't get to right now). The key is that the 'widget' you bind to an event (like left-Mouse in his example) get (x,y) pixel coordinates from the system which given appropriate information from par() can be translated into your data coordinates in just a few steps. James' site is a treasure chest for R and tcltk. Highly recommended. Hope this helps, Dirk -- 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://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] How to plot a small figure in a bigger one???
Hi jzhang10 wrote: Hi, I want to insert a small figure into a bigger plot. I saw people are doing this all the time, but I just could not figure out how to do it in R. Thanks for your help! Jinfeng Zhang There are a number of ways to do this in R; some pointers to get you started ... (i) par(new=TRUE) allows you to overlay several plots on the same page (ii) par(fig) and par(plt) provide fine control over where a plot appears on a page (iii) accurately locating subplots within a bigger plot may require some conversions between coordinate systems; par(usr), par(plt), ... can be used to obtain information on current coordinate systems (iv) you can have all the transformations done for you, by using the grid package to create viewports and the gridBase package to align them with a bigger plot - see the recent R News article Integrating grid Graphics Output with Base Graphics Output (http://cran.stat.auckland.ac.nz/doc/Rnews/Rnews_2003-2.pdf) Paul -- 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://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Virus vers expéditeur
Votre message : hi destiné à [EMAIL PROTECTED] contenait un virus. __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] print comment lines on `sink'ed files?
Hi, (B (BI am using many comment lines like this (B# a comment line (Bin a script. Is there any way to print (Bcomment lines when I use sink() function? (BOr, are there any plans to include as (Ban option to sink() function in the near (Bfuture? I would not want to use cat() (Bfunction as it will look messy in the (Boriginal script file. (B (BThanks! (B (BSI (B (B__ (BDo You Yahoo!? (BYahoo! BB is Broadband by Yahoo! (Bhttp://bb.yahoo.co.jp/ (B (B__ (B[EMAIL PROTECTED] mailing list (Bhttps://www.stat.math.ethz.ch/mailman/listinfo/r-help (BPLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] 3 little questions
See ?cor.test Also, rowSums(outer(unique(r1),r1,==)) gives the number of occurrences of the elements of unique(r1) --- Date: Sun, 01 Feb 2004 16:53:21 -0800 From: Siegfried.Macho [EMAIL PROTECTED] To: [EMAIL PROTECTED] Subject: [R] 3 little questions Dear R-helpers, 3 questions: 1. Is there a package that contains a routine for computing Kendall's W (coefficient of concordance), with and without ties ? 2. Is there a package that contains a routine for computing Goodman' s Gamma. 3. I there a simple method for computing the number of ties as well as their lengths within a vector fo ranks, e.g. r1 - rank(c(1, 3, 2, 3, 3, 2, 4)) gives: [1] 1.0 5.0 2.5 5.0 5.0 2.5 7.0 which contains 2 ties with length 2 and 3. Thanks in advance, S.M == Dr. Siegfried Macho Department of Psychology University of Fribourg Rue de Faucigny 2 CH-1700 Fribourg __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Stepwise Regression and PLS
--- Frank E Harrell Jr [EMAIL PROTECTED] wrote: On Sun, 1 Feb 2004 11:09:28 -0800 (PST) Jinsong Zhao [EMAIL PROTECTED] wrote: Dear all, I am a newcomer to R. I intend to using R to do stepwise regression and PLS with a data set (a 55x20 matrix, with one dependent and 19 independent variable). Based on the same data set, I have done the same work using SPSS and SAS. However, there is much difference between the results obtained by R and SPSS or SAS. In the case of stepwise, SPSS gave out a model with 4 independent variable, but with step(), R gave out a model with 10 and much higher R2. Furthermore, regsubsets() also indicate the 10 variable is one of the best regression subset. How to explain this difference? And in the case of my data set, how many variables that enter the model would be reasonable? In the case of PLS, the results of mvr function of pls.pcr package is also different with that of SAS. Although the number of optimum latent variables is same, the difference between R2 is much large. Why? Any comment and suggestion is very appreciated. Thanks in advance! Best wishes, Jinsong Zhao In your case SPSS, SAS, R, S-Plus, Stata, Systat, Statistica, and every other package will agree in one sense, because results from all of them will be virtually meaningless. Simulate some data from a known model and you'll quickly find out why stepwise variable selection is often a train wreck. --- Frank E Harrell Jr Professor and Chair School of Medicine Department of Biostatistics Vanderbilt University For the case of stepwise regression, I have found that the subsets I got using regsubsets() are collinear. However, the variables in SPSS's result are not collinear. I wonder what I should do to get a same or better linear model. Thanks! __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Stepwise Regression and PLS
--- Frank E Harrell Jr [EMAIL PROTECTED] wrote: For the case of stepwise regression, I have found that the subsets I got using regsubsets() are collinear. However, the variables in SPSS's result are not collinear. I wonder what I should do to get a same or better linear model. I think you missed the point. None of the variable selection procedures will provide results that have a fair probability of replicating in another sample. FH --- Frank E Harrell Jr Professor and Chair School of Medicine Department of Biostatistics Vanderbilt University Do you mean different procedures will provide different results? Maybe I don't understand your email correctly. Now, I just hope I could get a reasonable linear model using stepwise method in R, but I don't know how to deal with collinear problem. = (Mr.) Jinsong Zhao Ph.D. Candidate School of the Environment Nanjing University 22 Hankou Road, Nanjing 210093 P.R. China E-mail: [EMAIL PROTECTED] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] print comment lines on `sink'ed files?
SI [EMAIL PROTECTED] wrote: I am using many comment lines like this # a comment line in a script. Is there any way to print comment lines when I use sink() function? Or, are there any plans to include as an option to sink() function in the near future? I would not want to use cat() function as it will look messy in the original script file. Since you are using script files, you may not need to use R in an interactive session. One suggestion under unix would be: cunegonde:~ R --no-save --quiet test saved cunegonde:~ cat saved # a comment line i-1:10 outer(i,i,*) [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10] [1,]12345678910 [2,]2468 10 12 14 16 1820 [3,]369 12 15 18 21 24 2730 [4,]48 12 16 20 24 28 32 3640 [5,]5 10 15 20 25 30 35 40 4550 [6,]6 12 18 24 30 36 42 48 5460 [7,]7 14 21 28 35 42 49 56 6370 [8,]8 16 24 32 40 48 56 64 7280 [9,]9 18 27 36 45 54 63 72 8190 [10,] 10 20 30 40 50 60 70 80 90 100 # the end All commands including comments appear in the saved output. HTH, -- Philippe Glaziou Epidemiologist __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Stepwise Regression and PLS
Jinsong Zhao wrote: Do you mean different procedures will provide different results? Maybe I don't understand your email correctly. Now, I just hope I could get a reasonable linear model using stepwise method in R, but I don't know how to deal with collinear problem. What Dr. Harrell means (in part) is that stepwise regression leads to models that often overfit the observed data pattern--i.e. models that are not generalizable. More elaboration can be found here (including comments from Dr. Harrell): http://www.gseis.ucla.edu/courses/ed230bc1/notes4/swprobs.html Key quote: Personally, I would no more let an automatic routine select my model than I would let some best-fit procedure pack my suitcase. The bottom line advice here would be: don't use stepwise regression. Peter Kennedy, in A Guide to Econometrics (pp. 187-89) suggests the following options for dealing with collinearity: 1. Do nothing. The main problem in OLS when variables are collinear is that the estimated variances of the parameters are often inflated. 2. Obtain more data. 3. Formalize relationships among regressors (for example, in a simultaneous equation model). 4. Specify a relationship among the *parameters*. 5. Drop one or more variables. (In essence, a subset of #4 where coefficients are set to zero.) 6. Incorporate estimates from other studies. (A Bayesian might consider using a strong prior.) 7. Form a principal component from the variables, and use that instead. 8. Shrink the OLS estimates using the ridge or Stein estimators. Hope this helps. Chris -- Dr. Chris Lawrence [EMAIL PROTECTED] - http://blog.lordsutch.com/ __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] 3 little questions
From: Siegfried.Macho Dear R-helpers, 3 questions: 1. Is there a package that contains a routine for computing Kendall's W (coefficient of concordance), with and without ties ? Is that the same as Kendall's tau, as in cor(..., method=kendall)? 2. Is there a package that contains a routine for computing Goodman' s Gamma. Don't know, if you can find out by search on CRAN. 3. I there a simple method for computing the number of ties as well as their lengths within a vector fo ranks, e.g. r1 - rank(c(1, 3, 2, 3, 3, 2, 4)) gives: [1] 1.0 5.0 2.5 5.0 5.0 2.5 7.0 which contains 2 ties with length 2 and 3. Try table(), which gives you frequencies. HTH, Andy Thanks in advance, S.M == Dr. Siegfried Macho Department of Psychology University of Fribourg Rue de Faucigny 2 CH-1700 Fribourg Tel.: +41-26-3007635 Fax.: +41-26-3009712 http://www.unifr.ch/psycho/general/macho.htm __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Notice: This e-mail message, together with any attachments,...{{dropped}} __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] print comment lines on `sink'ed files?
Thanks for your help, Phillipe. You're right; I don't (Bneed to use sink() function. I was using R for windows in (Ban interactive mode and have totally forgotten about batch (Bmode. (B (BFor others' reference I opened up a command prompt and (Btyped in: (B (B Rterm --no-save --quiet test.R test.log (B (Band created test.log gave me the results I wanted. (B (B (BCheers, (B (BSI (B- (BSI [EMAIL PROTECTED] wrote: (B I am using many comment lines like this (B # a comment line (B in a script. Is there any way to print (B comment lines when I use sink() function? (B Or, are there any plans to include as (B an option to sink() function in the near (B future? I would not want to use cat() (B function as it will look messy in the (B original script file. (B (B (B (BSince you are using script files, you may not need to use (BR (Bin an interactive session. One suggestion under unix would (Bbe: (B (B (Bcunegonde:~ R --no-save --quiet test saved (Bcunegonde:~ cat saved (B # a comment line (B i-1:10 (B outer(i,i,"*") (B [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10] (B [1,]12345678910 (B [2,]2468 10 12 14 16 1820 (B [3,]369 12 15 18 21 24 2730 (B [4,]48 12 16 20 24 28 32 3640 (B [5,]5 10 15 20 25 30 35 40 4550 (B [6,]6 12 18 24 30 36 42 48 5460 (B [7,]7 14 21 28 35 42 49 56 6370 (B [8,]8 16 24 32 40 48 56 64 7280 (B [9,]9 18 27 36 45 54 63 72 8190 (B[10,] 10 20 30 40 50 60 70 80 90 100 (B # the end (B (B (BAll commands including comments appear in the saved (Boutput. (B (BHTH, (B (B-- (BPhilippe Glaziou (BEpidemiologist (B-- (B (B__ (BDo You Yahoo!? (BYahoo! BB is Broadband by Yahoo! (Bhttp://bb.yahoo.co.jp/ (B (B__ (B[EMAIL PROTECTED] mailing list (Bhttps://www.stat.math.ethz.ch/mailman/listinfo/r-help (BPLEASE do read the posting guide! http://www.R-project.org/posting-guide.html