Page 1 of 1

interpretation results from linear model

PostPosted: Thu Mar 03, 2011 1:07 pm
by qzx
Dear all,

I have a model which I log-linearized myself. I declare the model as linear by Model (linear) and get the results. But I don't know how to properly interpret the results. All the variables in the Model block are actually percentage deviation from the ss, so the moments reported are the variance of those percentages or the variance of the level of original variables? If the moments are the variance, autocorrelation, correlation of the percentages, how can I get the variance, autocorrelation, correlation of the levels?

best,

Zongxin

Re: interpretation results from linear model

PostPosted: Mon Mar 21, 2011 11:19 am
by SébastienVillemot
The moments reported by Dynare are indeed those of the percentage deviations.

If you want the moments of the original variables, the easiest way is to write the model in its orginal form (non-linear equations, variable in levels). Dynare nicely handles these situations.