[Limdep Nlogit List] A selection qeustion--a quick follow-up

[Fred Feinberg]

Christer.Thrane at hil.no wrote:

> Thanks everybody! Bottom line: Since I do not know anything about Bayesian
> statistics, I can't adjust my multinomial regression model for samlpe
> selection. This also means that my best approach probably is to go on as
> if selection is not a problem?

No! Don't give up the good fight (as they say), and leave an important
construct out of your model. There are MANY ways to estimate this model
classically... possibly even using LimDep, if you look at the Lee (1983)
paper. Write to me off-line if you want more information. [Prof. Greene would
know if it's possible to get LimDep to do it.] You can definitely do it using
quadrature, for example.

FF