Re: [Gretl-users] mle not finding correct maximum

On Sat, 5 Dec 2009, Davor Horvatic wrote: Attached is an example script which estimates a garch(1,1) model via mle with analytical derivatives. The script is pretty bare: suffice it to say that in order to compute the garch score analytically, we use the chain rule a lot, but I guess you're smart enough to figure out what's going on by yourself. In case you're wondering: the slight differences with the native command are due to the fact that in the C code we use a different (but asymptotically equivalent, in this case) method for initialising the conditional variance series h. Have fun! Riccardo (Jack) Lucchetti Dipartimento di Economia Università Politecnica delle Marche r.lucchetti(a)univpm.it http://www.econ.univpm.it/lucchetti