Dynare forums

Hi Johannes,

I have a question about Setting Parameters to their posterior mode after estimation. I estimated the model and now want to proceed to quantitative exercises. What is the correct way to proceed? First I just manually copied the values from estimaton and pasted them into the code. Is there a better way to do this?
And I'm also not quite sure what I should rather use - posterior mean or mode. Many papers write that they set Parameters to the posterior mode. In my case both statistics are quite Close and do not Change the implications of the model. What should be my argumenatation if I, for example, decide to use posterior mode?

Thanks in advance and best regards!

After Bayesian estimation, the parameters are set to the prior mean. This is what most people use. Regarding the quantitative exercises, it depends on what you want to do. Many exercises are usually conducted on the posterior distribution during estimation. If you just want to call stoch_simul after estimation, this can be done within one mod-file as the parameters stored in M_.params will be set to the posterior mean.