Dynare forums

When I do Bayesian estimation, it told me that "Prior distribution for parameter bhas two modes! Prior distribution for parameter xi has two modes!" (b: habit formation. xi: calvo stick).
Shall I ignore it? Or is there any method to deal with it?

If you look at the prior plots, you will see a mode at 0 and 1. Decrease the standard deviation of the beta distribution until you get a nice single-peaked distribution.

I see. Thanks a lot, Professor.
Warm regards,

Samson