Hello,
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
[email protected] Thank you.
Best,
Steve
*
* 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/
