[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: mprobit and ivlogit?
On 9/1/06, Clive Nicholas <Clive.Nicholas@newcastle.ac.uk> wrote:
I'd like to add to Rich's valuable tips on getting the best out of
-mprobit- by tackling Nishant's comment that, given his data and model, he
makes no distinction between -mlogit- and -mprobit- on 'theoretical'
I've been thinking that -mlogit- uses the sufficient statistic
principle that is characteristic of exponential family, and assumes
uncorrelated "errors" that have some sort of extreme value
distribution; while -mprobit- assumes random errors that have a
multivariate normal distribution with (possibly) correlated errors.
Integration over those errors can be performed analytically in the
case of -mlogit- with a simpler error structure, but has to be done
numerically in the case of -mprobit-. Any microeconometricians out
there (real ones unlike me) to comment on that?
* For searches and help try: