Solving a incomplete information model with a Kalman Filter


I would like to replicate the DSGE model from Andolfatto, Moran and Hendry (please see link below). The main feature of such model is a monetary policy regime subject to both transitory and permanent shocks (which cannot be fully observed by the agents). Agents use a kalman filter to solve this problem.

I am not sure that Dynare would be able to run this kind of model. However, as I am not an experienced user, I would like to hear from you if this impression is right or wrong.

Thanks in advance … s-rational


This can be done in Dynare, but is somewhat involved. Blanchard, L’Huillier, Lorenzoni (2013) have solved something like this with Dynare 3.65. Implementing their model in Dynare 4 is still on my to-do list.
Essentially, you have to append the model by the agent’s signal extraction problem and solve for the coefficients of this problem externally, before then passing this back into Dynare via a steady state file. You could try to contact Patrick Hürtgen who has used Dynare 4 as far as I know for Hürtgen (2014); Consumer misperceptions, uncertain fundamentals, and the business cycle (
Particularly his appendix might be useful.

Thank you for your response Professor Pfeifer. I will follow your suggestions.

Dear amuller

Do you have any news over getting the model from Professor Huertgen?

Unfortunately not…