Posteriors in different versions of dynare

I have been running a program on v3.6. The posteriors are very flat and the mode is a long way from the mode from optimization

Previously it had seemed ok on version 3.4 and that seems still to be true.

Also ran it on 3.5 and tends to be soemwhere between 3.4 and 3.6

Mode Optimum values are different in all cases.

What has changed re MH and optimzation between the various versions

I am also finding that posteriors estimated by v6 are very flat (all parameters have flat posteriors) resulting in very high variances while the same model estimated under v5 seems to return more “normal” results. Has the sampling algorithm changed?

You are most likely falling victim of the bug with the penalty in the objective function that appears in versio 3.062. I hope to provide a new version soon. The SVN version of version 3 is corrected. Use the download instructions for version 4, replacing dynare_v4 by dynare_v3

Best

Michel