Running a counterfactual simulation in Garcia-Cicco et al (2010) model code

[malikshuk]

Hello, Is there an easy way to run a counterfactual simulation with some of the estimated smoothed shocks being shut down? If this issue has already been discussed I would appreciate a link. The model file I am working with is for replicating Garcia-Cicco et al (2010). Thank you M

[JohannesPfeifer]

You need to use the simult_-function and select the smoothed shocks you want, starting from the initial condition. It should be similar to https://github.com/DynareTeam/dynare/blob/master/tests/kalman_filter_smoother/compare_results_simulation/fs2000.mod

[malikshuk]

Thank you for the suggestion. I tried t adjust my file accordingly but got the following error message:

_Cell contents reference from a non-cell array object.

Error in JPfeiferRBCinEMEsMS (line 348) ex=[ex oo.SmoothedShocks.(M_.exo_names{shock_iter})];

Error in dynare (line 223) evalin('base',fname) ;_

I am using Dynare 4.5.3 and unfortunately unsure how to deal with this problem. I am guessing it some sort of compatibility problem. Thank you M

[JohannesPfeifer]

@malikshuk The link above works for the unstable. In Dynare 4.5, the code is https://github.com/DynareTeam/dynare/blob/4.5/tests/kalman_filter_smoother/compare_results_simulation/fs2000.mod

[malikshuk]

Thank you for the link! It did not work immediately but once I removed the extension .Mean in lines 121 and 129 it seems to work. Somehow the extension .Mean works neither in Dynare 4.5 nor in unstable versions. Thanks again. Malik

[JohannesPfeifer]

@malikshuk Whether the mean-field exists or not depends on whether you run ML or Bayesian estimation and whether the MCMC was run.

[malikshuk]

Got it, thank you.