a gibbs sampler would be very useful for alternative analysis.
Hm, this looks like a candidate for a contributed function package. For the methods in gretl I'm also not sure why it would be really needed (read: there is no Bayesian stuff in gretl). Of course having stuff like this would be nice, but this strikes me more as a catchword without being more specific.
So closing this for now.
Sign up for the SourceForge newsletter:
You seem to have CSS turned off.
Please don't fill out this field.