Dynare forums

Good day everyone,

I am running a stochastic simulation with all the initial values equal to zero. The Steady State Results are all zero for all the variables but for some of them the irf graph does not get back to zero! which for me looks incompatible with the numerical results. I have tried it with the different irf's but still does not work. Any idea why?

Cheers

There are several possibilities

• Your IRFs are just hugely persistent and you need thousands of periods to get back to steady state
• Some variables have unit root. In this case, their IRF is permanent and they do not go back. You can check this by looking at the eigenvalues displayed by check
• You are simulating your model at higher order and the deterministic steady state is not equal to the ergodic mean.