Permanent shock in stochastic simulation

Stochastic simulations
1

dynare.org/manual/index_18.html

In deterministic simulation, we specify a permanent shock by the use of initval and endval.

But how to specify a permanent shock in stochastic simulation?

Thanks!

2

Even though you have a permanent shock, you should type your model in stationary terms, otherwise it cannot be solved. I am not sure if that helps…

3

I am also wondering : how to specify a permanent shock in stochastic simulation? Is there a good example?

4

Hi rosa,

z = z(-1) + zg;
zg=rho_z*zg(-1) + eps_z;
5

What @cmarch provided is a very general example of a unit root process with an autocorrelated shock. Generally, you would simply need to have a shock follow a random walk, i.e. the process needs a unit root.