Dynare forums

Hello everyone,

I am writing for the following problem: I am trying to compute an optimal policy problem in Dynare in a deterministic setting (e.g under perfect-foresight).
Unfortunately it seems that the command “ramsey_model”, the one meant for optimal policy under perfect foresight, is not working in the current version of Dynare and it will come out with the version 4.5 (btw is it known the date of release?).

Now, I was looking for a solution avoiding me either to rely on Lopez-Salido’s code (which seems to be incompatible with the current version of Dynare and Matlab) or to compute and type planner’s focs by hand.
The solution it came to my mind was to run the command ramsey_policy first so to trigger the creation of the m. files relative to the dynamic and static model (containing optimal policy focs), and after that calling the routine simul.m in order to perform deterministic simulation over this model.

In this respect I meant to ask: Do you think is this procedure correct? Or is there any further issue I should be aware about?

Many thanks in advance,

Mattia

Dear Mattia,
which Lopez-Salido code are you referring to?
Note that the ramsey_model command should already be available in the current Dynare unstable version.
Also, as you seem to be the first user trying to use this new functionality, you might want to contact Michel Juillard directly. I am sure he will help you out.

Dear Professor Pfeifer,
Thank you very much for your answer. Just few points more:

1. The code I was referring is the one from Levin (sorry that was I typo), get Ramsey. It seems to me that it is not compatible with the current version of Dynare and Matlab. Can you confirm?

2. I will definitely contact Michel Juillard, thanks. For the time being, do you think the solution I came up with can anyway work? (the solution was: run the command ramsey_policy first so to trigger the creation of the m. files relative to the dynamic and static model, containing optimal policy focs, and after that calling the routine simul.m in order to perform deterministic simulation over this model.)

Thank you very much,

Mattia Ricci