Dererministic Shock

[Lupin]

Hi guys, quick question.

I have a deterministic model and I have to simulate a specific path for the exogenous variables. I can do that by simulating a sequence of permanent shocks using as a starting value the outcome of the previous simulation, but then the model is not solved under perfect foresight since at date t agents expect the exogenous variable to be at level t forever, while at t+1 it will increase again. Is there a way to specify a path for the exogenous variable so that agents are not constantly surprised every period?

Thanks

[rteconomics]

No, if the shock is an exogenous variable agents will always be 'surprised'. Unless you specify the shock as a news shock. In that case agents are not surprised when the shock materializes but when the shock is announced.

[Lupin]

Thanks rteconomics. How do I specify a news shock?

[rteconomics]

TWell, the easiest way is this:

Code: Select all

a = rho*a(-1) + e(-1);

in this way the shock is 'announced' at (t-1) but materializes at time t. Agents react at (t-1) and not at t.