3rd order for big model

Stochastic simulations
1

Dear all,

I have a concern about working with risk/uncertainty shock with a big model (multisector model). As a common practice, the 3rd order is required. In the past, I noticed it takes a lot of resources to do it and NLMA toolbox can provide a shortcut for the IRFs. I wonder if there is any innovation from DYNARE to improve the computational constraints. I am quite sure 3rd order is hard to have for more than 100 equations model.

Thank you so much.

2

No, there is nothing. You can only profile your code and see where the bottlenecks are. Sometimes, different solvers for the Lyapunov equation help.