I have a case-control dataset with 530 observations, of which 137 cases and
393 controls. The observations are at the level of the households, but
households are grouped in villages (15), so I am interested in running a
random effect model, with a two-level specification. Specifically, I am
interseted in a random effect logit model since my outcome is binary.
Is the xtlogit procedure adequate to estimate my model, given the case-control
nature of the data? Are 15 villages enough to estimate level-2 variance in a
model that uses Gauss-Hermite quadrature?
Thanks in advance for the suport.
Manuela De Allegri
University of Heidelberg
Germany
*
* 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/
