similar to: time series analysis

Greetings, I am trying to install R (0.63.2) on a Digital Unix 4.0b Alpha Server 2100A using gcc 2.8.1 and f77 v 0.5.2.3. It seems to compile OK. However, when I try to run R I get the following message: R : Copyright 1999, The R Development Core Team Version 0.63.2 (January 12, 1999) R is free software and comes with ABSOLUTELY NO WARRANTY. You are welcome to redistribute it under

R ran find on our Sun until it was stolen. I have now set up R on one of our Dec Alphas OSF1 V4.0 386. R-0.49 configures builds and runs fine except that the glm command crashes R. R-0.60.1 configures builds and runs until a cis entered. I think the problem is the floating point conventions but before I get in over my head a question Has anyone used glm successfully on Alphas OSF1 V4.0 386?

I am new to R so may well have missed the point somewhere. I would like to use an exponential error in my generalized linear model. It seems natural to restrict the Gamma family to do this ( and as one might in GLIM) by specifying the scale. This does not seem possible in R . Have I missed something? Sorry to raise such a trivial point but I am keen to specify the scale G.Janacek

I have a probably very naive question about the R anova/aov functions: I've found in several text books with descriptions of anova procedures that, in nested anova (specifying the model "X~A/B", B being a factor specifying treatments, and A groups of samples within each treatment), the mean square for "among treatments" is divided by the mean square "within treatment

Hello R experts, I haven't found anything like the S function 'varcomp' as described in W.N. Venables & B.D. Ripley's 'Modern Applied Statistics ...' for R; does something comparable exist for R, or is planned for future releases? More generally, are there libraries with post-anova test procedures, like Student-Newman-Keuls? Or do those of you who frequently use

Hello R-Experts, 2 questions: does anybody have example functions for determining the power of, say, a t-test or a single factor anova with fixed or random factors? and: is there an implementation of Cochran's test for heterogeneity of variances (i.e., is there a frequency distribution of Cochran's C)? Many thanks for your attention ... Ulf (please CC: me, I'm not on the

Help ! I am a graduate student in statistics, nearing the end of my first semester, and recently discovered R, but cannot figure out how to install it on my Mac. I am currently running OS 10.1.1, and know next to nothing about Unix or anything related. I would like to install this program on my Powerbook, as it come highly recommended, so I can use it for a class next semester. Any help

Hello out there, my first try to install a package (e.g. ctest from CRAN) didn't succeed; looking at the error messages I think it must be a configuration problem of my machine, but as I'm not very experienced I would be glad if someone could give me a hint ... R's version is 0.63.0, my i386-Linux is kernel 2.0.35. I hadn't had problems to compile R, and R itself seems to work.

Dear List, After my laptop was stolen I (eventually) replaced it with a G4 laptop running Mac OS 9.1 . This also meant I updated to the current Mac binary 1.3.1. This is pretty wonderful except for one problem. I normally save R figs as eps files using the Mac OS and then read them into Latex documents. This no longer works as the legends overwrite the diagrams. Saving as ps from R also does

Excuse me, if the following is FAQ or nonsense, I'm new to R (and the list)! After trying to import a textfile of 184897 Bytes, R (v0.63.0 on Linux 2.0.35/DLD 5.4) gives the following error message: > read.table ("db.out") Error: heap memory (1953 Kb) exhausted [needed 150 Kb more] > Is there a possibility to increase

I have a similar question, I know about --vsize, but I use R under Emacs using ESS. I know there is an easy way to call R from ESS with command line options but I don't remember how. I can't find help on this in ESS documentation or in R documentation. Could someone please remind me how this is done. ------------------------------------ | Robert Denham | |

Hi Troels, Its in the incoming directory: ftp://ftp.ci.tuwien.ac.at/incoming Try this for a nice demo: > library(timeslab) > arma(alpha, beta, x, 2, 4, 3, 0.2,500, 30,0) Bernd On 01-Jan-99 Troels Ring wrote: > Hi Johannes Wuebben and a happy new year. > Where did you actually put the timeslab ? > Yours > > Troels Ring, MD > Department of Nephrology

Hi, I guess the answer is negative but I thought I would ask just the same: is it possible to use any of the S+ toolboxes (ARCH and wavelets, more particularly) with R? Thanks, Olivier Lefevre Union Bank of Switzerland New York, NY PS: Please cc me any reply as I am not on the list (at least not yet).

Hello again, A different but related question to my last one: Does anyone know if one can easily estimate state-space models using ML and the kalman filter using R? I would be especially interested in a relatively flexible function that would allow for estimation of hyperparameters, or could be made to do so. Thanks Michael J. Roberts Resource Economics Division, PMT USDA-ERS 202-654-5557

hello I read "Practical Time Series" (Gareth Janacek; 2001) and they presented e.g the smoothing functions msmooth(x,k) or the bivariate function crosscorr(x,y,k), but both didn't work on my machine. I only load the ts library, is another library necessary or did this function change since 2001? Is there a more recent and detailed manual for ts? thanks, cheers Martin -- Martin

Hi, I have just seen Stefan Steinhaus' web page : http://www.uni-franfurt.de/~stst/ncrunch.html I think it would be nice to include "R" as well. I have taken Forrest Young's email on stat-lisp list and changed the stuff for "R" :) Here it is: (someone please check this so we can also send it to Stefan Steinhaus.

Thanks to all for the input. rgentlem at jimmy.harvard.edu said: >If you really want help -- no, I just thought I'd waste some bandwidth. :-) >If on the other hand you >just want some reassurance that Splus is indeed a commerical package oh gee, thanks, I wasn't sure about that. When I had said: >> R can't handle files named x. If I can use Splus, should I just

Hello everybody, shame on me if I have overlooked something (CRAN, StatLib), but I think I've searched carefully. Maybe it's just too obvious to see for me (happens frequently). Is a function available that extracts multiple comparison of means from objects produced by "aov()" ? Sorry if the term is not correct, I translated from german word by word ("Multiple

I noticed that congigure reported libpng as "not found" although a recent version exists. Further looking into this issue told me, that the test failed because the test from R.m4 AC_CHECK_LIB(png, png_create_write_struct, [ BITMAP_LIBS="${BITMAP_LIBS} -lpng -lz" AC_DEFINE(HAVE_PNG) ], , ${LIBS}) ], AC_MSG_RESULT([no])) ]) did

I am running some simulations which involve fitting a glm with Binomial error ( in fact Bernoulli) and logistic link. Running R 0.63 I have no problems. However I need more speed so migrated to our UNIX boxes which run Version 0.90.1 (December 15, 1999) Here I get an error Error in (if (is.empty.model(mt)) glm.fit.null else glm.fit)(x = X, y = Y, : y must be univariate unless binomial