Monday, January 02, 2006
st: Question: obtaining robust standard errors from Poisson regressions
I'm writing with a question about obtaining robust standard errors from a Poisson regression. Specifically, I would like to know if it's possible to obtain robust standard errors that are valid under any arbitrary assumption about the conditional variance (as discussed, for example, on p. 650-651 in Wooldridge 2002).
I looked at both the "poisson" and the "vce_option" entries in the Stata reference manual. It's not clear to me whether any of these options can produce standard errors that are valid under any assumption about the conditional variance. I presume that by default, the poisson command displays estimated standard errors that are valid only under the assumption that the conditional variance equals the conditional mean.
I would greatly appreciate any advice. I can be reached at firstname.lastname@example.org Thank you.
* * 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/
Links to this post: