Re: [Gretl-users] bivariate probit (in a loop)
On Wed, 22 Oct 2014, Artur Bala wrote:
Dear Allin,
There's another strange behavior out there: I run exactly the same script
several times and the results are always different.
Well, may be the problem is the buglet you pointed out yesterday???
Besides, a new error message came out this time.
Here's attached the output.
There's no doubt your likelihood function is misbehaved here (rho near 1
is worrying) and what you're seeing are numerical problems. Could you
re-run a few times with the --verbose switch on and send us the output?
-------------------------------------------------------
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
-------------------------------------------------------