s-news
[Top] [All Lists]

Re: glmD (poisson)

To: jose Bartolomei <surfprjab@hotmail.com>
Subject: Re: glmD (poisson)
From: Frank E Harrell Jr <f.harrell@vanderbilt.edu>
Date: Tue, 03 Feb 2009 07:18:47 -0600
Cc: s-plus news <s-news@lists.biostat.wustl.edu>, Thomas Dupont <charles.dupont@vanderbilt.edu>
In-reply-to: <BAY105-W2209AB9A4A7AAF4D15D0F1B6C20@phx.gbl>
References: <BAY105-W2209AB9A4A7AAF4D15D0F1B6C20@phx.gbl>
User-agent: Thunderbird 2.0.0.19 (X11/20090105)
jose Bartolomei wrote:
Dear S-xperts, Can you help me interpret the following error. Problem in addOffset4ModelFrame: No data to interpret as logical value:
 if(offsetVarname %nin% names(newdata)) {
 newdata[[offsetVarname]] <- rep(offset, length = nrow(newdata))
 warning(paste("offset variable set to", paste(format(offset), collapse
   = " ")))
This happens when I run: ibrary(Hmisc, T)
library(Design, T)

ddist<- datadist(ERVarray4Regression,
  adjto.cat=c(Gender="first", hic="first", Year="first", ageG="first"))
options(datadist="ddist") ERV.glm1<-glmD(ERVclaims~Gender + offset(logPop) , family=poisson(link=log), data=ERVarray4Regression)

summary(ERV.glm1)
I do not get error when use glm Thanks in advance,
Jose

I tried a simpler version of this on the latest version of Design on R and did not get this problem. Please provide a minimum example that results in the error, that we can reproduce, using simple simulated data (providing the simulation code) and tell us the version of S-Plus you are using and the version or date of the Design package. Also ask Tibco to update Design using the latest R package on CRAN, as well as to update Hmisc.

Frank

--
Frank E Harrell Jr   Professor and Chair           School of Medicine
                     Department of Biostatistics   Vanderbilt University

<Prev in Thread] Current Thread [Next in Thread>
  • glmD (poisson), jose Bartolomei
    • Re: glmD (poisson), Frank E Harrell Jr <=