Page 1 of 1
theoretical moments differ from simulated ones
Posted:
Wed May 26, 2010 6:39 pmby marcin_j
Hi there, I'm just simulating a very basic rbc model. Everything seems to be in order but the moments obtained in simulations are completely different than the theoretical ones. Could anybody give me a hint why it may be so? mod file attached if needed. Thanks a lot!
Re: theoretical moments differ from simulated ones
Posted:
Wed May 26, 2010 7:26 pmby jpfeifer
Sorry, but I do not understand what are you doing in your mod-file. You do not compute theoretical moments as you always specified that periods is not 0 (see Dynare manual). If you are comparing the first and the second stoch_simul command in your file, you are comparing HP-filtered with unfiltered data. Moreover, Dynare does not standardize the standard deviations by the mean of the respective variable, so std(y)/mean(y) has not corresponding expression in Dynare.
Re: theoretical moments differ from simulated ones
Posted:
Wed May 26, 2010 9:10 pmby AssiaEzzeroug
Hi,
I think it may have something to do with the simulation order because to my knowledge, only in order 1 simulated moments- that are then computed from simulated data- should reproduce theoretical moments (computed from the theoretical expression) while If I well understood from your mod-file, you are trying to compare them in order 2.
Best