similar to: Documentation Bugs (PR#1618)

Seems to be a day for finding peculiar little things. There is an inconsistency in the behavior of lm vis a vis glm: > x <- rnorm(15) > y <- 1 + 10*x + rnorm(15) > z <- as.factor(rep(c("A","B","C"),rep(5,3))) > xyz <- data.frame(x,y,z) > fit.lm <- lm(y ~ x + z, data=xyz, subset=(z != "C")) > fit.glm <- glm(y ~ x +

Hi Peter and others, If it helps, I wrote a small function glm.scoretest() for the statmod package on CRAN to compute score tests from glm fits. The score test for adding a covariate, or any set of covariates, can be extracted very neatly from the standard glm output, although you probably already know that. Regards Gordon --------------------------------------------- Professor Gordon K

I'm posting this for two reasons: one is to see if anyone has a better way of solving the problem or suggestions for improving my existing approach; and the other is to show what I'm currently doing in case anyone else might find it useful. The background is that I've been using Sweave for several years now to produce class notes, and I sometimes include quite a lot of raw R input and

I'll start with my apologies to Martin for sending those last two of message to the list owner (the last one he forwarded was actually sent a week or two ago and must have been lost in the ether somehow). This is basically due to fast copy/paste in emacs without paying much attention to what I'm doing. Lack of sleep is my excuse, but I don't know if that goes very far with the

How exactly should I go about turning off annotation when running Stangle (Rtangle) with "R CMD Stangle myfile.Rnw"? Ideally I would like to be able to turn annotation off and on for individual code chunks, or maybe better, to annotate only named chunks. Are either of these things easily done? Is there some way to figure out this sort of thing without reading through the source code

Can someone explain to me in what way the new (dpqr)gamma parameter can be interpreted as a rate (when shape != 1)? The only gamma rate that I am aware of is the hazard rate given by dgamma/(1-pgamma), the log of which is returned by my hgamma function (event library). Jim -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-help mailing list -- Read

This might more properly be filed as a bug report, but ... I came upon the following problem while trying to dyn.load a library of Fortran code into R. I'm running RedHat 7.1 on a Pentium III laptop, with R version 1.3.1 (latest rpm from CRAN) and gcc/g77 version 2.96. My library has a number of Fortran subroutines that have underscores in their names for readability. By default g77

{again a message that was sent to owner-r-help (which is me, currently) why on earth ???!??!? reply to R-help or the original sender Brett Presnell; } I'm teaching a categorical data analysis course this term, and a minor "problem" has resurfaced that I have often thought about before. This applies equally to Splus I suppose, but my undergrads aren't using Splus. It

Is there any reason that rstandard.glm doesn't have a "pearson" option? And if not, can it be added? Background: I'm currently teaching an undergrad/grad-service course from Agresti's "Introduction to Categorical Data Analysis (2nd edn)" and deviance residuals are not used in the text. For now I'll just provide the students with a simple function to use, but I

Here's a simple example: > x <- rnorm(20) > y <- 1 + 10*x + rnorm(20) > plot(y ~ x, subset=2*(1:10),xlab="In",ylab="Out") Error in length(x) == l : comparison (1) is possible only for vector types The problem seems to be in this section of the code for plot.formula: if (!missing(subset)) { s <- eval(m$subset, data, parent.frame()) l

Hello, I have found a problem with using the glm function with a gamma family. I have a vector of data, assumed to be generated by a gamma distribution. The parameters of this gamma distribution are estimated in two ways (i) using the glm() function, (ii) "by hand", using the optim() function. I find that the -2*likelihood at the maximum found by (i) is substantially larger than that

I'm trying to use the vcd package to produce mosaic plots for my class notes, written in Sweave and using the LaTeX's beamer document class. For projecting the notes in class, I use a dark background with light foreground colors. It's easy enough to change the defaults for R's standard graphics to match my color scheme (using the fg, col.axis, col.lab, col.main, and col.sub

Dear All, We are about to purchase the second edition of Agresti's "Categorical Data Analysis" (my old copy of the first ed. of that wonderful book is falling apart). I would appreciate suggestions about other comparable books which, if possible, have examples using R/S code (instead of SAS). Thanks, Ram?n -- Ram?n D?az-Uriarte Bioinformatics Unit Centro Nacional de

I apologize in advance if this is the wrong forum for this report/request, and for the fact that I have not read the code for ftable.formula in any detail. >From reading the documentation for ftable.formula, I expected that the following two calls to ftable would produce the same results: data(UCBAdmissions) ftable(UCBAdmissions, row.vars = "Dept", col.vars = c("Gender",

{why do I get more and more messages sent to "owner-r-devel" or "owner-r-help" ?? "owner-foo" by convention (even RFC ???) is the (set of) maintainer(s) of mailing list "foo".. Please *do* post to "R-help" , "R-devel", and *NOT* to "owner-...." ! Martin Maechler = current owner-r-... ------- start of forwarded

Hi I am using R for logistic regression and finding it very useful. However, I wondered if anyone could point me to any course or notes on this subject using R. All help most welcome. Stephen -- Internal Virus Database is out-of-date. Checked by AVG Anti-Virus. [[alternative HTML version deleted]]

I have just committed some code to the r-devel branch to implement the Rao efficient score test. This is asymptotically equivalent to the LRT, but there is some indication that it might have better properties in smaller samples since it is based more directly on the distribution of the sufficient sums under the null hypothesis (e.g., if you have a divergent fit to the model under the alternative,

--fupGvOGOQM Content-Type: text/plain; charset=us-ascii Content-Description: message body and .signature Content-Transfer-Encoding: 7bit There's a simple bug in the handling of the start and offset arguments in glm and glm.fit. The bug exists in the latest development version of R (version information below), but it appears that glm.R has not been touched much lately, so the bug affects at

Dear R-help, just for the record a summary of responses to my problem: > On 04/05/2010 9:41 AM, Rainer Scheuchenpflug wrote: > a student of mine tries to use the Windows-Rconsole with screen > reading software (she is blind), and cannot access the command line (Menus are ok). - Duncan Murdoch (Murdoch.duncan at gmail.com) suggested as preliminary workaround to run RTerm in a command

Hello As I'm beginning to write functions, I decided to have my personnal R library named MyStuff located in /home/fred/R.Lib/MyStuff/ The library file itself is at /home/fred/R.Lib/MyStuff/R/MyStuff evrything seems ok since > library() gives : Packages in library `/usr/lib/R/library': KernSmooth Functions for kernel smoothing for Wand &