[Gretl-users] Re: query regarding random effect tobit regression
On Fri, 12 Feb 2021, Sven Schreiber wrote:
Am 12.02.2021 um 09:28 schrieb Riccardo (Jack) Lucchetti:
> On Fri, 12 Feb 2021, mandeep kaur wrote:
>
>> I have also run simple tobit model. I provides the output. But when I
>> tried
>> to run random effect panel tobit regression, it didn't work.
>
> That's strange. Could you please share your data with us so that we
> can see what's going on?
Jack, but apart from this problem, why is the --opg option used for the
tobit initialization at all? AFAICS only $coeff is grabbed, and that
doesn't depend on opg, or does it?
I believe it's used to make initialisation a little faster, but I really
can't remember :D
-------------------------------------------------------
Riccardo (Jack) Lucchetti
Dipartimento di Scienze Economiche e Sociali (DiSES)
Università Politecnica delle Marche
(formerly known as Università di Ancona)
r.lucchetti(a)univpm.it
http://www2.econ.univpm.it/servizi/hpp/lucchetti
-------------------------------------------------------