Page 1 of 1

bayestopt_

PostPosted: Thu Oct 17, 2013 5:51 am
by Dr. En Ami
What does the t and opt (or topt) stand for in bayestopt_, as found in the dynare code?

Some of its fields seem to duplicate those of options_. Is there some general principle as to which of the two values in options_ and bayestopt_ has precedence or which one would apply when?

Does [M_.fname '_optimal_mh_scale_parameter.mat'] have precedence over an mh_scale specified in the estimation command?

Re: bayestopt_

PostPosted: Tue Oct 22, 2013 9:54 am
by jpfeifer
I have no idea what the t is for. The opt is for options. Note that bayestopt_ only govern estimation and options_ all of Dynare. Thus, the former is more local in scope and should usually dominate in estimation. Note also that there is no way for users to specify fields of bayestopt_ so that they don't have to deal with which option is used.

Regarding, [M_.fname '_optimal_mh_scale_parameter.mat']: unless you are using mode_compute=0, the file will always be deleted and the one specified in mh_scale will be used.