I wonder that here nobody else seems to be interested in estimating such models. They should fit to a number of different applications.


Am 16.11.2013 um 17:40 schrieb Pindar <pindar777@gmail.com>:

In this paper a quite detailed analysis of the the problem is given:

On page 919 is says that Fixed Effect Poisson can be estimated by a conditional MLE that evolves to the multinomial logit.
Since the latter model can be estimated in gretl it would be great to take this approach. However, I don't get the ends meet in order to make this work.

The non plus ultra would be a --fixed-effects with for the poisson command, but a MLE or conditional MLE UDF also has its charm.

12.11.2013 18:30, Pindar:
So, after reading the appropriate literature I now know, that in principle Poisson FE could be estimated by just including the unit dummies.
However, there are too many of them in my data set and now I'm stuck with implementing this log-likelihood


I think it wont work with 'mle' because of the 'within sums over t periods' and needs an approach like in felogit.

06.11.2013 11:54, Riccardo (Jack) Lucchetti:
On Wed, 6 Nov 2013, Pindar wrote:

I found Jack's felogit.gfn on the server. I guess there is no feprobit because of the incidental parameters problem, or?

Yes. Moreover, if you have a panel dataset you can use the --random-effects options to the probit command to obtain RE probit estimation via Gaussian quadrature. By the way, your post made me realise this is still undocumented, although it's been in for a while and I've even used it in a paper! I'll try to update the docs asap.

  Riccardo (Jack) Lucchetti
  Dipartimento di Scienze Economiche e Sociali (DiSES)

  Universit√† Politecnica delle Marche
  (formerly known as Universit√† di Ancona)


Gretl-users mailing list