Questions about Bayesian estimation?

Dear Professor Pfeifer,

I have encountered some questions when doing the Bayesian estimation. Would you please give some advice? Any reply will be appreciated. Thank you very much.

(1) When I set the “mode_compute=4”, the result says “POSTERIOR KERNEL OPTIMIZATION PROBLEM”, Why does this happen? Is there anything wrong with the code or data? or I just have to chage the value of priors and try again?

(2) I could get a result by setting mode_compute=6, but it is not ideal. What can I do to get a good result ? Adjust the value of priors or the setting of "mh_drop"and " mh_jscale " ? or is there any other ways?

This indicates a problem with the mode-finding. It can be caused by a variety of issues. The most common one is the data treatment/observation equation. The mode_check-plots often provide a good first indication. But without files it is hard to tell.