Learning code?

Dear all,

I used to have the same post but not in the right section of the forum I guess.

I was wondering if anyone could send to me an example Dynare code that performs Bayesian estimation of a DSGE model but in an adaptive learning setting (e.g. CG learning)?

Thank you very much in advance for your help!

Hi George

I ma not really sure, but I think that learning is not implementable in Dynare yet. If you look at the DynareTimeTable (cepremap.cnrs.fr/DynareWiki/DynareTimeTable) they are planning to do that by the end of the year.

Paolo

Hello,

though a few years later, I have the same problem like George. I’m looking for a way to implement adaptive learning in dynare. Did you found a way how to estimate your model with adaptive learning in dynare? Is this in the meantime possible? If so how?

I don’t know how deal with the time varying coefficients you face under learning. Is there a way in dynare to deal with it, or another work around so to implement learning in a DGSE model. I was wondering if maybe someone has an example how to to estimate adaptive learning within a Bayesian DGSE model. Can someone help me? Any help is appreciated!

Thank you very much!!!

I am also interested in this issue.
Please someone help us!

Sorry, but at the current stage, there are no functions for learning in Dynare. The manual should be corrected soon.

You would have to write a Matlab loop around resol() and updating the parameters (the learning) in Matlab in the loop.
Your best shot may be the code on aeaweb.org/articles.php?doi=10.1257/mac.4.2.65
You should also check whether it is contained in Volker Wieland’s macromodelbase, which has an adaptive learning team, see macromodelbase.com/editor