Comparing the empirical and theoretical moments

Hi everybody,
In the case of the nonlinear model , how can i compare the empirical and theoretical( simulated) moments( for example the MEAN or VARIANCE ) ?
I mean, the values of theoretical moments are small numbers, but the empirical moments are not.

Then you are doing something wrong. You are supposed to compare the same objects in the model and the data.

I suppose I didn’t ask my questions properly.In my model, production is only a function of working time y=f(n) and because the working time is normalized to 1, the value of the theoretical moments of the y and other variables the steady state will be small numbers , but the value of empirical moments of variables calculated based on actual data extracted from national accounts are big numbers. What can be done to compare the empirical and theoretical moments?

Due do constant returns to scale, the mean is meaningless in any case. For second moments, the standard procedure is to compare percentage (i.e. log) deviations from the steady state so that the scaling does not matter.

1 Like