Dynare forums

Hi there,

I try to simulate a model and the mod file passed the rank condition. But then the simulation shows a lot "NaN" for theoretical moments and finally displays error information below. Really appreciate if anyone can help me with this problem. I attach my mod file below as well.

??? Subscripted assignment dimension mismatch.

Error in ==> disp_th_moments at 103
z(:,i) = diag(oo_.gamma_y{i+1}(i1,i1));

Error in ==> stoch_simul at 134
disp_th_moments(oo_.dr,var_list);

Error in ==> simul at 431
info = stoch_simul(var_list_);

Error in ==> dynare at 132
evalin('base',fname) ;

The "NaN" in the moments correspond to the non stationary variables of your model. There are a lot of them, since you have three unit roots.

The crash is due to the fact that all your variables are either non stationary or constant (zero variance). This is a very special case which is so far not handled nicely by Dynare. Note that it is impossible to compute autocorrelations in that case. In future versions, Dynare will display a more intelligible error message.

Best,