On Thu, 18 Oct 2007, yinung CYCU wrote: While in principle it's perfectly legitimate to estimate a garch model without a mean equation, this is not possible at present in gretl via the "garch" command. In order to ensure that this requirement is fulfilled, the inclusion of a constant in the mean equation is presently hardwired. The source would have to be modified quite substantially to accommodate models with no mean equation; on the other hand, if one wanted to estimate models with exogenous variables other than the constant in the mean equation, the amount of work required would be much less and probably doable in a shorter timespan. However, we're very close to releasing 1.6.6 and I don't think it would be wise to embark in such a project right now, also considering that substantial enhancements to the garch code are planned for the future anyway. Of course, this is open source, so if you feel like writing some C code yourself, feel free to ask for help. In the meantime, if you absolutely need this, you can always use the mle command (suboptimal but reliable). Riccardo (Jack) Lucchetti Dipartimento di Economia Università Politecnica delle Marche r.lucchetti(a)univpm.it http://www.econ.univpm.it/lucchetti