Everyone,
I'm going through some diagnostics on some models using xtgee (poisson and
negative binomial family). The Hardin & Hilbe book includes a discussion of
using DFBETA values to check for influential observations but the book's
companion do files don't have the code to calculate the values. I've also
not been able to find an implementation of this on the Stata list archives
or by searching the net. Just curious if anyone here has calculated DFBETAs
after running GEE models and, if so, if they might be willing to share some
code examples?
Thanks,
Tim
------------------------------------------------------------------
Tim Quigley
Ph.D. Candidate
Department of Management and Organization
Smeal College of Business
The Pennsylvania State University
416A Business Building
University Park, PA 16801
[email protected]
------------------------------------------------------------------
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
