Stochastic depreciation rate of capital

[rteconomics]

Hi guys,
I'm working to a mickey-mouse model of a closed economy in which the depreciation rate of capital is stochastic (+ a traditional TFP shock).
I have two issues:

1) When the shock is to the capital stock the model runs fine. However, when the shock is the TFP shock I don't get the IRFs... weird isn't it? I attach the code so maybe someone can take a quick look.

2) Another weird issue is about the parametrization. In particular the weird parameter is 'phiy', the persistence of the TFP shock. When this is 1, in the static model all the equations clear (meaning the residuals are zero). But when it's set to a lower level there are convergence issues to the steady-state (which is bizarre because the shock itsefl is set to zero in the steady state).

Any idea? THANKS!
RT

[jpfeifer]

Your model has a unit root, which is responsible for the issues.

[rteconomics]

Hi,
Oh I see.. even if I don't understand why the model runs fine when considering the shock to the capital stock..
In any case, you're right. I need to close the model first.
Ciao and many thanks!
RT