If you prefer using R, then this is probably the way to go; however, if you are interested in what's βunder the hoodβ in Stata, you can always see what happens in much the same way as in R, using
set trace on
to find out which code works (or load tr from SSC) or using
viewsource xtnbreg.ado
to see the actual code executed by xtnbreg .
If you are interested in how Stata computes results in xtnbreg , a detailed discussion of the likelihood function on the [XT] page on pages 367-370 with links included.
eric.a.booth
source share