Simulated variables after estimation of a DSGE model

[vk2020]

Dear friends,

I would like to retrieve the simulated series of endogenous variables after I have performed bayesian estimation. I have 4 series: inflation, output gap, interest rates and stock prices. I tried to use the stoch_simul (periods = integer) command after estimation and then compared the simulated variables to actual data. The result is very obscure. The series are not even close.

Can anyone help?
Many thanks.

Best,
VK

[kyri82]

You mean that when you plot them, they are not similar at all?

[vk2020]

No, not at all. I was hoping that the simulations would closely follow the path of actual variables.

[flumas22]

Dear vk2020,

If these are series that you have used in estimation (i. e. observed variables), use simply command filtered_vars which plots estimated variables using one-sided Kalman filter. The Kalman filter estimates of the observed variables can be given the same interpretation as the fitted values of a regression.

For example:

estimation (datafile=?, ................, filtered_vars, ..........)

Best, Jan

[vk2020]

Dear Jan,

Thanks for the suggestion. I tried to look at the plot of the actual vs filtered variables. The filtered variables almost look like the smoothed/updated varibales. I was hoping I could get the simulated variables. Any suggestions?

Best,
VK