Learning mechanism in Dynare


Could someone give me some hints about writing adaptive learning mechanism, such as constant gain learning, in Dynare?

Thank you very much

I’m looking for some hints too. There is a Learning Toolbox written for Matlab. Has Dynare 4 incorporated these features yet?

I try to estimate a dsge model with adaptive learning expectation using dynare 4, the problem is to write a time-varying low of motion into the kalman filter,
someone could help me please.

There is no feature in Dynare for modeling learning. Learning, in most cases, is characterized by changing values of the coefficients. The only way to do it is to write a Matlab loop around stoch_simul() that takes care of revising the value of the parameters. You need to know some Dynare internals in order to do so.