Implicit observables in Bayesian Estimation


I’m estimating a big DSGE model with financial accelerator using bayesian methods. Clearly I have less observables than the number of variables in my model, also because some of the data that I have are not really good and they make worse the estimation when I introduce them.
In any case I wanted to know if it could be possible to get the implicit behavior of the variables which are not considered as observables in the estimation but on which I have data in order to compare them.



Hi Cristiano, You may compare the smoothed variables (obtained with the Kalman smoother) with your data. Look at the manual, there is an option (of the estimation command) to trigger the computation of the smoothed variables.