[R] Checking the assumptions for a proper GLM model

Jay josip.2000 at gmail.com
Thu Feb 18 18:44:40 CET 2010


Well, yes and no. Obviously I was not asking for the complete recap of
a all the theory on the subject. My main concern is finding readily
available CRAN functions and packages that would help me in the
process. I've found the UCLA site to be very informative and spent a
lot of time ther the last couple of days. However, their section on
using R for validating the assumptions is very lacking. Naturally
links like google.com and amazon.com will eventually get me there, but
if somebody have other recommendations, I would be very fortunate to
get even more help.

BR,
Jay

On Feb 18, 7:01 pm, David Winsemius <dwinsem... at comcast.net> wrote:
> At one time the "answer" would have been to buy a copy of Venables and  
> Ripley's "Modern Applied Statistics with S" (and R), and that would  
> still be a sensible strategy. There are now quite a few other R-
> centric texts that have been published in the last few years. Search  
> Amazon if needed. You seem to be asking for a tutorial on general  
> linear modeling (which if you read the Posting Guide you will find is  
> not a service offered by the r-help list.)  Perhaps you should have  
> edited the link you provided in the obvious fashion:
>
> http://www.ats.ucla.edu/stat/R/
>
> Perhaps one of these pages:http://www.ats.ucla.edu/stat/R/dae/default.htm
>
> The UCLA Statistics website used to be dismissive of R, but they more  
> recently appear to have seen the light. There is also a great amount  
> of contributed teaching material on CRAN:
>
> http://cran.r-project.org/other-docs.html
>
> ... and more would be readily available via Googling with "r-project"  
> as part of a search strategy. Frank Harrell's material is in  
> particular quite useful:
>
> http://biostat.mc.vanderbilt.edu/wiki/Main/StatComp
>
> --
> David.
>
> On Feb 18, 2010, at 8:32 AM, Jay wrote:
>
>
>
>
>
> > So what I'm looking for is readily available tools/packages that could
> > produce some of the following:
>
> > 3.6 Summary of Useful Commands (STATA: Source:
> >http://www.ats.ucla.edu/stat/Stata/webbooks/logistic/chapter3/statalo...)
>
> >    * linktest--performs a link test for model specification, in our
> > case to check if logit is the right link function to use. This command
> > is issued after the logit or logistic command.
> <snipped>
> > and performs nonlinearity test.
>
> > But, since I'm new to GLM, I owuld greatly appreciate how you/others
> > go about and test the validity of a GLM model.
>
> > On Feb 18, 1:18 am, Jay <josip.2... at gmail.com> wrote:
> >> Hello,
>
> >> Are there any packages/functions available for testing the  
> >> assumptions
> >> underlying assumptions for a good GLM model? Like linktest in STATA
> >> and smilar. If not, could somebody please describe their work process
> >> when they check the validity of a logit/probit model?
>
> >> Regards,
> >> Jay
>
> David Winsemius, MD
> Heritage Laboratories
> West Hartford, CT
>
> ______________________________________________
> R-h... at r-project.org mailing listhttps://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guidehttp://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.



More information about the R-help mailing list