Page 1 of 1

Deterministic permanent shock after 5 periods

PostPosted: Mon Nov 04, 2013 3:21 pm
by Philipp
Hi there,

please help me out with a little problem I have here:
I want to simulate a simple deterministic permanent shock that occures not in period 1 but say in period 5.
How can this be done?
I used initval; and endval; to specify the permanent shock, but as I interpret the output the permanent shock happens in period 1, which is not exactly what I would like.

Thanks for your help.
Philipp

Re: Deterministic permanent shock after 5 periods

PostPosted: Mon Nov 04, 2013 4:16 pm
by jpfeifer
You are solving under perfect foresight. Agents know that in period 5 there will be a permanent shock and react today. What you want to have is a permanent shock at time 1. Before that, all variables will simply be at their initial conditions.

Re: Deterministic permanent shock after 5 periods

PostPosted: Tue Nov 05, 2013 7:13 am
by Philipp
Dear jpfeifer,
thanks for your prompt reply. I always appreciate your help very much.
So I guess what I wanted is only a plot problem, which rplot cannot handle: I wanted something like for t=-10:0 to be at the initial steady state and from t=1:... to be at the new level. In this way, the drop (increase) after the shock would be visible.
I will need to compute this in a separate m.file, no big deal.
Best, Philipp

Re: Deterministic permanent shock after 5 periods

PostPosted: Tue Nov 05, 2013 9:40 am
by jpfeifer
In principle, you should be able to access oo_.dr.ys before running simul. This should provide you with the initial condition.