similar to: R-beta: problem with gee() with singletons

On Tue, 1 Apr 1997, Jim Lindsey wrote: > Neither setting environment nor command > line change memory size as given by gc(). However, if I give too small > or too large a value on the command line, it tells me that it is > invalid and is ignoring it!?? I repeat that when I recompiled changing > the default size, gc() changed its values. Am I misunderstanding > something? I

There was a question earlier today, on either r-help or r-devel, about calling R code from FORTRAN (I deleted it accidentally, so I don't know which list it belonged to). Anyway, there is a workaround to call R functions from FORTRAN which is used in the "integrate" library. It was invented for S by Mike Meyer. The FORTRAN code needs to be written to call a fixed C function and

rbind() does something strange to dimnames R : Copyright 1997, Robert Gentleman and Ross Ihaka Version 0.50 Beta (April 1, 1997) R is free software and comes with ABSOLUTELY NO WARRANTY. You are welcome to redistribute it under certain conditions. Type "license()" for details. > test1 <- data.frame(time= c(4, 3,1,1,2,2,3), + status=c(1,NA,1,0,1,1,0), + x= c(0,

If you subscript a data frame using the names of the columns and there is more than one column with the same name only the first one is found. eg mm<-data.frame(1:10,log(1:10)) names(mm)<-c("(offset)","(offset)") mm[,"(offset)"] # only returns first column mm[,"(offset)"]<-rep(1,10) # only sets first column This *is* compatible with S, but I

The hat() function, returning the diagonal of the hat matrix, currently works only on matrices and doesn't work with weights. Could we make it return lm.influence(x)$hat when inherits(x,"lm")? Thomas Lumley ------------------------------------------------------+------ Biostatistics : "Never attribute to malice what : Uni of Washington : can be adequately explained by :

Is there a good reason not to allow matrix extents to be zero? I have been writing a function which fits glms under certain linear constraints. I fit the model with a reduced set of variables and then transform back to the original set. This involves operating on submatrices of the covariance matrix and subsets of the coefficients. Sometimes these subsets are empty. A zero-length subset of the

Both of these used to work and seem useful and harmless: R> matrix(1,ncol=1)%*%c(1,2) Error in matrix(1, ncol = 1) %*% c(1, 2) : non-conformable arguments R> matrix(1,ncol=1)*(1:2) Error: dim<- length of dims do not match the length of object Thomas Lumley -----------------------------------------------------+------ Biostatistics : "Never attribute to malice what : Uni of

Both of these used to work and seem useful and harmless: R> matrix(1,ncol=1)%*%c(1,2) Error in matrix(1, ncol = 1) %*% c(1, 2) : non-conformable arguments R> matrix(1,ncol=1)*(1:2) Error: dim<- length of dims do not match the length of object Thomas Lumley -----------------------------------------------------+------ Biostatistics : "Never attribute to malice what : Uni of

Any hints on how to do a significance test using gee? Is the "robust z" the thing to use? There is apparently another implementation of gee called yags (http://www.biostat.harvard.edu/~carey), but I can't figure out how to install it. I think it does significance tests. -- Jonathan Baron, Professor of Psychology, University of Pennsylvania Home page:

Hi, I'm attempting a comparative analysis using the function compare.gee, package (ape), which uses phylogeny as a correlation matrix in gee (package gee), in R version 2.2.0 on a Windows XP PC. I'm looking to model the relationship between a continuous explanatory variable and a binary response variable for 600 species, taking into account the phylogeny of those species. Here's

Hi, I am fitting a GEE model using gee R package, but I am not sure how to get OR and its CI? Could anyone give me some hints? Here are some output: > gee.obj <- gee(Affection~Sibsex+Probandsex,id = FAMID,family = binomial,corstr = "independence",data =seldata) Beginning Cgee S-function, @(#) geeformula.q 4.13 98/01/27 running glm to get initial regression estimate

I'm not sure if the gee package is still actively maintained, but I for one find it extremely useful. However, I've come across a few infelicities that I'm hoping could be resolved for future versions. Hope it's okay to list them all in one post! They are: (1) AR(1) models don't fit when clustsize = 1 for any subject, even if some subjects have clustsize > 1. (2) If the

Dear all, I need to fit a gee model with an auto-regressive correlation structure and I faced some problems. I attach a simple example: ####################################################### library(gee) library(geepack) # I SIMULATE DATA FROM POISSON DISTRIBUTION, 10 OBS FOR EACH OF 50 GROUPS set.seed(1) y <- rpois(500,50) x <- rnorm(500) id <- rep(1:50,each=10) # EXAMPLES FOR

Greetings, I have a very, very, simple research question. I want to predict one dichotomous variable using another dichotomous variable. Straightforward, right? The issue is that the dataset has two issues causing some complications for me. 1) The subjects are not independent -- they are sibling pairs. Every person in the dataset has a sibling in the dataset. This needs to be treated a

Hi, I would like to fit a GLM model with GEE on clustered data. I tried to use gee in the GEE package on a twin data set. All cluster are of size 2. I removed the missing data and ordered by IDENTIF2 first. library(gee) mod.pc <- gee(Y ~ X1 + X2 , id = IDENTIF2, family = binomial, corstr = "unstructured", data = na.omit(df)) gives the following result : Beginning Cgee S-function,

Behavioral science data is often collected from nested structures (students in schools, in districts, etc.). This can produce nonindependence among responses from individuals in the same groups. Consequently, researchers are advised to model the nested nature of the data to avoid biases in SE estimates. Failing to account for nonindependence can lead to SE estimates that are too large or too

I'm using R 1.9.1 on suse server v9 enterprise with the gee package version 4.13-10. I have code that runs in an automated script. It uses the gee function from the gee package. The script is run quite often without error. I have a problem where the script locks up R when calling this function (it starts execution and never finishes). I was able to track down the offending data and determine

Dear R-help, I was recently wanting to use GEE with the negative binomial "family". It seems that this is lacking in the otherwise excellent implementations of the GEE methodology ( packages: gee, yags, geepack). I would have thought it a simple step to allow the creation of a family, i.e providing the link function (log mu) and the variance function (mu + mu^2/theta) , assuming theta

Hi, having used both the gee pacakge and the geepack package, i am unsure of how to interpret the results. Here are the results from the geeglm function from the geepack package > gee2<-geeglm(data$erythema~data$product, data = data, id=subject, > family=binomial, corstr="independence") Warning message: In model.response(mf, "numeric") : using

Dear List: I found an error when I called the 'gee' function. I cannot solve and explain it. There are no errors when I used the 'geeglm' function. Both functions fit the gee model. The project supervisor recommends me to use the 'gee' function. But I cannot explain to him why this error happens. Would you help me solve this problem? I appreciate your help. In