Conditional forecast starting from steady state

Estimation ML/Bayesian estimation
1

Hi,

I am trying to get something similar to Bayesian IRFs but I want to set the value of an endogenous variable for a determined number of periods starting from the SS (for example an interest rate of 1% above SS for 5 periods). Ideally, everything would be centered at 0.

I tried to use conditional forecast right after estimation for that. But I could not find a way of starting the forecast from the SS. As I understand the forecast starts at the end of the estimation filtered values.

I managed to get standard IRFs using simult_ but setting the shock value, not the endogenous variable. which is not exactly what I want.

Is it possible to do that?
Thanks!!

2

Did you set parameter_set=calibration?

3

No, I want to use the posterior to do that.
The issue is that the forecast does not start in SS.
A minor issue is that the graph not centered at 0, but I guess I can fix that plotting with Matlab.

4

After estimation, the parameter set will be set to the posterior mean. When you set parameter_set=calibration you should therefore get the right parameter set, but the simulation will start at the steady state.