Calibrated smoother for the law of large numbers

General DSGE Modeling
1

Dear all,

I have one question about calibrated smoother.

To observe the mean effect, we conduct simulations, and just as the law of large numbers suggests using 1,000,000 simulations rather than 100 simulations for accurate results.

How can we apply the law of large numbers in a calibrated smoother to obtain reliable outcomes?

Thank you so much for your help in advance!

Best wishes,
R.

2

I am not sure I understand the question. For the calibrated smoother, you fix the parameter vector. Hence, there is no parameter uncertainty. For the disturbance uncertainty there are analytical formulas that let you compute the mean states at order=1. No simulations are necessary.

3

Thanks a lot! Happy new year!!