Theoretical Moments and Data

Stochastic simulations
1

Dear all,

Sorry for asking this question, it most probably has been addressed many times in forum before. I have estimated my model and simulated it at the calibrated posterior mean. Dynare produces the theoretical moments, which I want to compare with the data to assess the model’s fit. I have used the command:

stoch_simul(order=1, irf=20, irf_shocks = (eps_r, eps_zh, eps_p, eps_mh), graph_format = pdf, conditional_variance_decomposition = [4 8]) variable1 variable2 …

The only problem is that I have quadratically detrended the data. So I’m not sure how to compare it to the actual data. I know that we usually want to compare HP filtered theoretical moments with the HP filtered data. However, how would I do this in my case? I should also mention my model in entered in Dynare nonlinearly and observables are mean zero.

Many thanks for your help,

Robert

2

Your observation equation describes the mapping between the data and the model variables. You should thus compare the moments of the model implied observables with the data you feed in. Put differently, your model provides you with series for the quadratically detrended data. You can easily compare those from the model to the actual quadratically detrended data

1 Like
3

Many thanks Johannes!