Page 1 of 1

params_bounds

PostPosted: Thu May 11, 2006 7:15 am
by twlxv
Dear all,

I have a question on "estimated_params_bounds", I use this command with one of the parameter that I estimate, assigning the interval 0.00001, inf to it, but after the MH iteration, the confidence interval for that parameter still includes negative value, does anyone why this happen?

Thanks,

Regards,

Wing

Re: params_bounds

PostPosted: Thu May 11, 2006 7:57 pm
by MichelJuillard
estimated_params_bounds is only used as a help for computing the posterior mode, it isn't used during MH iterations. It is only for avoiding that the optimizer gets itself lost in regions where the mode can't reasonably be.

It isn't a device to do constrained estimation. If you want to restrict a parameter to positive numbers, use a prior such as gamma_pdf or inv_gamma_pdf

Kind regards

Michel