Estimating model with optimal commitment policy

Estimation
1

Hello,

I am trying to estimate a parameter with GMM after solving the model with an optimal commitment policy. I get this error message.

“method_of_moments: The steady state at the initial parameters cannot be computed.”

I am not sure what I am doing incorrectly. I don’t get any error messages when I solve for the optimal commitment policy and only get this error message when I estimate the model. Could you help me figure this out? Does it have something to do with having to change the steady state values as the value of the parameter that I am trying to estimate changes? If so, how should I go about fixing this?

Thank you.

The files that I am using are attached below.

data.xls (25.5 KB)
full_com_mp.mod (2.9 KB)
setup.m (2.4 KB)
steadystate.m (1.8 KB)

2

I think I found the solution. The steady state is numerically solved in a different file, but I define the steady state values as parameters in the mod file. When I estimate the model, even though the parameter values change, the steady state values do not change since they are not solved for again in the steady state block.

To fix this, I created a separate steady state file that numerically solves the steady state values.

3

Yes, that explanation makes sense. Steady states always need to be conditional on the structural parameters.