Is it possible to estimate optimal policy rule on real data?
Yes, if you write the mode with ramsey_policy, you can then put estimation instruction. You will be estimating the parameters of the model on the maintained hypothesis that the policy maker was following an optimal policy.
Best
Michel