How to model a structural break in a DSGE model

Dear Johannes,
Thank you very much for your previous help, I am grateful.
I have three questions.

  1. How to model a structural break in the DSGE model? for example a change in a smaller number of parameters after a known structural break
  2. After a structural break, given that some parameter values have changed, the steady state value of the DSGE system will also change right?
  3. If we would like to model a phenomenon that the steady state value of coal consumption has decreased after a structural break, the only thing we can do is to assume that some parameters associated with the calculation of the coal consumption’s steady state have shifted after the structural break?
    Thank you very much and look forward to hearing from you.
    Best regards,

Hi Jesse.

  1. The short answer is: Dynare does not handle structural breaks in DSGE models. You might want to take a look at RISE Toolbox, which solves and estimate regime-switching models. Nevertheless, depending on what you are trying to do, there are some workarounds to incorporate structural breaks. For instance, level shifts can be achieved by means of unit-root processes. Also, in a perfect foresight environment, it is easy to incorporate breaks in non-linear models (treat parameters as endogenous variables).

  2. It depends. Some parameters pin down the steady state, while others don’t.

  3. It depends. What caused the steady state to shift? Are you thinking about an endogenous or exogenous change in coal consumption level?

Professor Johannes might want to add to my answer.

1 Like

I would also refer you to Estimation with deterministic break - #4 by cjj245

1 Like