Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Grouped Logit question


From   Jhilbe@aol.com
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Grouped Logit question
Date   Thu, 22 Apr 2004 22:12:20 -0400

Why would you want to use -glogit- rather than -glm- with the family(binomial) and link(logit) and the proportional denominator listed as indicated in the help file for -glm-? The fit and residual statistics are appropriately adjusted, and you can use a variety of estimating alternatives. 

Joe Hilbe

> Revealing my ignorance of the mathematical proofs, I'm struggling to 
> get a grip on exactly what the Grouped Logit function does and how it 
> should be used appropriately. Specifically I wonder about the use of 
> count vs. proportion variables on both the left and right hand side.
> 
> the command is (of course)
> 
> glogit depvar denominator list_of_covariates
> 
> Also, I assume that what this does (in layman terms) is essentially 
> weight each row by the relative size of that observation compared to 
> the size of the other observations. In my case I have observations by 
> county and running the regression for the full US... i.e. I 
> have some 
> 3200 observations representing all use households.
> 
> Thanks in advance,
> 
> Fredrik
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index