I have a model (run with glm) that has a factor, type. Type has two levels, "general" and "regional". I am trying to get estimates (and SEs) for the model with type="general" and type ="regional" using fit.contrast but I can't get the syntax of the coefficients to use in fit.contrast correct. I hope someone can show me how to use fit.contrast, or some other method to get estimate with SEs. (I know I can use the variance co-variance matrix, but I would rather not have to code the linear contrast my self from the coefficients of the matrix) Thank you, John My model: model=glm(events~type,family=poisson(link=log),offset=log(SS),data=data) Model details:> summary(data$type)general regional 16 16> levels(data$type)[1] "general" "regional"> contrasts(data$type)regional general 0 regional 1 I have tried the following syntax for fit.contrast fit.contrast(model,type,c(1,0)) and get an error: Error in `[[<-`(`*tmp*`, varname, value = cmat) : no such index at level 1> fit.contrast(model,type,c(0,1),showall=TRUE)and get an error: Error in `[[<-`(`*tmp*`, varname, value = cmat) : no such index at level 1> fit.contrast(model,type,c(1,-1),showall=TRUE)and get an error: Error in `[[<-`(`*tmp*`, varname, value = cmat) : no such index at level 1> fit.contrast(model,type,c(0))and get an error: Error in make.contrasts(coeff, ncol(coeff)) : Too many contrasts specified. Must be less than the number of factor levels (columns).> fit.contrast(model,type,c(1))Error in make.contrasts(coeff, ncol(coeff)) : and get an error Too many contrasts specified. Must be less than the number of factor levels (columns). John David Sorkin M.D., Ph.D. Professor of Medicine Chief, Biostatistics and Informatics University of Maryland School of Medicine Division of Gerontology and Geriatric Medicine Baltimore VA Medical Center 10 North Greene Street GRECC (BT/18/GR) Baltimore, MD 21201-1524 (Phone) 410-605-7119 (Fax) 410-605-7913 (Please call phone number above prior to faxing) [[alternative HTML version deleted]]

> On Oct 22, 2017, at 6:04 AM, Sorkin, John <jsorkin at som.umaryland.edu> wrote: > > I have a model (run with glm) that has a factor, type. Type has two levels, "general" and "regional". I am trying to get estimates (and SEs) for the model with type="general" and type ="regional" using fit.contrast?fit.contrast No documentation for ?fit.contrast? in specified packages and libraries: you could try ???fit.contrast? Perhaps the gmodels function of that name?> but I can't get the syntax of the coefficients to use in fit.contrast correct. I hope someone can show me how to use fit.contrast, or some other method to get estimate with SEs. (I know I can use the variance co-variance matrix, but I would rather not have to code the linear contrast my self from the coefficients of the matrix) >I'm having trouble understanding what you are trying to extract. There are only 2 levels so there is really only one interesting contrast ("general" vs "regional") , and it's magnitude would be reported by just typing `model`, and it's SE would show up in output of `summary(model)`. I'm thinking you should pick one of the examples in gmodels::fit.contrast that most resembles your real problem, post it, and and then explain what difficulties you are having with interpretation. -- David.> Thank you, > > John > > > My model: > > model=glm(events~type,family=poisson(link=log),offset=log(SS),data=data) > > > Model details: > >> summary(data$type) > > general regional > 16 16 > >> levels(data$type) > [1] "general" "regional" > >> contrasts(data$type) > regional > general 0 > regional 1 > > > I have tried the following syntax for fit.contrast > > fit.contrast(model,type,c(1,0)) > and get an error: > Error in `[[<-`(`*tmp*`, varname, value = cmat) : > no such index at level 1 > > >> fit.contrast(model,type,c(0,1),showall=TRUE) > and get an error: > Error in `[[<-`(`*tmp*`, varname, value = cmat) : > no such index at level 1 > > > >> fit.contrast(model,type,c(1,-1),showall=TRUE) > and get an error: > Error in `[[<-`(`*tmp*`, varname, value = cmat) : > no such index at level 1 > > >> fit.contrast(model,type,c(0)) > and get an error: > Error in make.contrasts(coeff, ncol(coeff)) : > Too many contrasts specified. Must be less than the number of factor levels (columns). > >> fit.contrast(model,type,c(1)) > Error in make.contrasts(coeff, ncol(coeff)) : > and get an error > Too many contrasts specified. Must be less than the number of factor levels (columns). > > > > > > > > > John David Sorkin M.D., Ph.D. > Professor of Medicine > Chief, Biostatistics and Informatics > University of Maryland School of Medicine Division of Gerontology and Geriatric Medicine > Baltimore VA Medical Center > 10 North Greene Street > GRECC (BT/18/GR) > Baltimore, MD 21201-1524 > (Phone) 410-605-7119 > (Fax) 410-605-7913 (Please call phone number above prior to faxing) > > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code.David Winsemius Alameda, CA, USA 'Any technology distinguishable from magic is insufficiently advanced.' -Gehm's Corollary to Clarke's Third Law