Interpretation of Policy Function and Simulation


I have a function .mod file for my model. As I run my model with a second order solution in the output I get the policy function coefficients. I was wondering how to correctly simulate the model’s data through this policy functions.

Also, as I understand, these policy functions relate past values of state variables and the realizations of shocks. However, there is not a way to relate the model in a foward looking fashion relating state variables with also the expected value of some variables?


  1. For the first part, simulating the model, you shoud use the simult_-function. An example is at where the model is simulated forward with user-provided shocks

  2. The whole point of finding a solution in state-space form is to rely only on information available at time t. The expectations of future variables are solved for by replacing them by a function of the state variables.