what if questions to evaluate alternative forecasts
Posted: Thu Jun 24, 2010 12:05 amDear all,
Suppose I estimate a model where i add the forecast option in the same command (results are saved as usual). Not content with the forecast, I ask what would happen if the parameters of the Taylor MP rule change, for example.
One obvious way to respond it is to reestimate with Taylor parameters calibrated at the new values. But this takes too long. I don't want to do that (as long as the inicial state is the same) how would you suggest to tune these parameters to affect the forecasts ? (the other parameters remain at the post mean if you do mode_estimate=0).
what i've done so far did not result (I've intervened dynare_estimation_1.m forcing the parameter vector to a value). Any help or insight is much appreciated.
Best,
Jorge
Suppose I estimate a model where i add the forecast option in the same command (results are saved as usual). Not content with the forecast, I ask what would happen if the parameters of the Taylor MP rule change, for example.
One obvious way to respond it is to reestimate with Taylor parameters calibrated at the new values. But this takes too long. I don't want to do that (as long as the inicial state is the same) how would you suggest to tune these parameters to affect the forecasts ? (the other parameters remain at the post mean if you do mode_estimate=0).
what i've done so far did not result (I've intervened dynare_estimation_1.m forcing the parameter vector to a value). Any help or insight is much appreciated.
Best,
Jorge