Re: [Gretl-users] Quadratic form in log-likelihood (CONT'D)
On Sat, 10 Mar 2018, Alecos Papadopoulos wrote:
So I would really appreciate any idea why the following does not work
(it gives immediately a numerical derivative value of exactly zero for
the rho correlation parameter irrespective of starting value, which is
suspicious, and stays there for ever,
Well, in the code fragment you sent the parameter rho is never used in the
mle loop, so its derivative is 0. I guess you should move the statement
matrix R2 = {1,rho; rho,1}
inside the mle loop.
-------------------------------------------------------
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
-------------------------------------------------------