Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: pseudo R square after svy logistic regression


From   "JVerkuilen (Gmail)" <jvverkuilen@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: pseudo R square after svy logistic regression
Date   Fri, 30 Nov 2012 00:16:17 -0500

On Thu, Nov 29, 2012 at 7:56 PM, Ameya Bondre
<ameyabondre.jhsph@gmail.com> wrote:
> Hello,
>
> I have a survey design and I wanted to know how I can explain the
> variability in the data by logistic regressions, since my response and
> predictor variables are all binary, and because the log likelihood and
> pseudo R2 are not provided after using svy:logit y x1 x2...
>
> Is there any other way of getting an equivalent for R2 in such a
> situation? I am concerned if the Odds Ratios and confidence intervals
> in the logistic regressions would have any meaning without an R2 or
> its equivalent. The data has been obtained through stratified
> one-stage cluster sampling.

Generally when Stata doesn't report something, it means that in the
best guess of their statisticians, it's not a meaningful number. There
are legitimate disagreements about that but when you're dealing with
weighting as happens in survey data, I bet that's the situation. You
could always calculate it by hand using McFadden's formula if you have
to. In binary regression pseudo-R^2 is rarely meaningful anyway.
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


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