similar to: gee p values

Hello, I am interested in running generalized estimating equation models in R. Currently there are two main packages for doing so in R, geepack and gee. I understand that even though one can obtain similar to almost identical results using either of the two, that there are differences between the packages. The paper that introduces the geepack package (

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, List, I am using the package 'gee' downloaded from CRAN several days ago. I tried to do a 2 degree freedom contrast but couldn't get any result. It must be so obvious but I just couldn't get it. Any hint would be greatly appreciated. Here is a sample code: set.seed(123) mydf <-

Dear List I am using the gee package and want to specify my own correlation structure using the corstr(fixed) option. Can anybody advise me how exactly I need to define the structure I want? Do I create a matrix myself? How can I get the command to realise it needs to use my own matrix? Many thanks Lauar

Hi R-Help. I am working on a data set with a 3-level nested structure. I have individuals nested in households and multiple observations on each individual. I assume that the individuals inside a given household are correlated and that the individuals are correlated with themselves over time. The data is not balanced. I have computed a GLMM with logit link function and two random normal

We are currently analyzing data on children clustered in day care-centers (DCC). We have tried to use geepack and gee libraries to estimate an overall incidence rate for absences (=number of absences/risk time) by specifying geese(number.absences ~ offset(log(risktime)), id=day.care.id, family=poisson("log"), data=dcc, corstr="exch",

Hi A quick question. Standard errors reported by gee/yags differs from the ones in geeglm (geepack). require(gee) require(geepack) require(yags) mm <- gee(breaks ~ tension, id=wool, data=warpbreaks, corstr="exchangeable") mm2 <- geeglm(breaks ~ tension, id=wool, data=warpbreaks, corstr="exchangeable", std.err = "san.se") mm3 <- yags(breaks ~

Hi, I'm using 'geepack' to run Generalized Estimating Equations. I'm aware that I can use anova to compare two models, but would it be possible to test QIC on R? It seems that there were similar questions a couple of years ago, but the question has not been answered yet. I'd appreciate if someone could show me the code! Thank you, Sachi [[alternative HTML version

Hi I'm having some difficulty understanding my output (below) from GEE. the person who wrote the program included some comments about the '3-th term gives diff between hyp/ox at time..', and after created an L vector to use for a WALD test. I was wondering if someone could help me understand the GEE output, the programmers comment, how L was determined, and its use in the WALD

All - I would like to run a repeated measures analysis using non normal error (glm). Is "gee" the appropriate package? Else? Any real life examples using longitudinal data? Thanks, Keith

Dear all, Does anybody happen to have written some GEE source code in Fortran to solve a Generalized Estimation Equation Model? Or kindly point out me a good starting point. Thanks. Zhu Wang SCHARP

Hello, R exits every time I want to fit a GEE with AR-1 correlation structure. It just closes itself without any error. My Data has about 3300 observation (using the big data with about 20 000 observation it doesn't work at all). I'm a Windows 2000 User of R 2.0.1 but it was the same problem on a MacOSX. Is it my fault? Thank you for your help! Johanna Brandt

Hi, I want to fit a GEE with a negative binomial distribution. I have uesd already a poisson glm and then neg binommial to deal with alot of dispersion. In my neg binomial residuals i have some patterns so i have implemented a GEE, but only with a poisson family as i couldnt with neg binomial. However the residual patterns in fact look worse here. When i try and put neg binomial family it

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,

Hi there, I am using the geepack package and wish to test if my data (family=poisson) is overdispersed so that I may adjust the "scale.value" if necessary. Is there a specific function or method I should be using? Thanks for the advice, Bruce _____________________________ Bruce Catton MSc Candidate Forest Sciences University of British Columbia Vancouver, British

Hello, Is it possible to derive Pan's QIC (2001 Biometrics 57:120) from either a fitted gee() object in the gee package or from a geese() fit in the geepack package? If so, would anyone be kind enough to provide me with code to do so? I realize that QIC is part of the output from yags() but I would like to use one of the other functions. Thanks. Richard

Dear R users, I fitted a GEE model using the function 'geese' (or 'geeglm') with user defined correlation matrix. I want to get the var-cov matrix of the regression coefficients. But the output provides only limited information. I would be very much thankful if you could kindly let me know how to get it..since I am struggling lot getting this. Thanks -- View this message in

A bug (misfeature) has been found in the S gee library (and thus in the R gee library). The problem, which is shared by nearly all gee implementations, involves the calculation of working correlations when some clusters have only one observation. For compatibility reasons nearly everyone uses the computing formula from the first SAS macro by Karim, rather than the formula from the original GEE

Hi, How to perform a stepwise selection of the best models for gee? Why can't step() do this job? Thanks. Sincerely Yours, Jinn-Yuh Guh, M.D. Dept. of Internal Medicine Kaohsiung Medical College 100 Shi-Chuan 1st Road Kaohsiung, Taiwan FAX: 886-7-312-2810 e-mail: jyuh at mail.nsysu.edu.tw -------------- next part -------------- An HTML attachment was scrubbed... URL:

Is there any way to perform stepwise Poisson regression with GEE in R? (I know, and agree that stepwise procedures are not best statistical practice. I have developed a model with 10 independent variables, my client want me to include 50 other independent variables. I, of course, don't what to do this. I am hoping that I can demonstrate that adding additional variables does not make a material