Working Paper: NBER ID: w26310
Authors: Enrique G. Mendoza; Sergio Villalvazo
Abstract: We propose a simple and fast fixed-point iteration algorithm FiPIt to obtain the global, non-linear solution of macro models with two endogenous state variables and occasionally binding constraints. This method uses fixed-point iteration on Euler equations to avoid solving two simultaneous nonlinear equations (as with the time iteration method) or creating modified state variables requiring irregular interpolation (as with the endogenous grids method). In the small-open-economy RBC and Sudden Stops models provided as examples, FiPIt is used on the bonds and capital Euler equations to solve for the bonds decision rule and the capital pricing function. In a standard Matlab platform, FiPIt solves both models much faster than time iteration and various hybrid methods. The choice of functions that FiPIt iterates on using the Euler equations can vary across models, and there can be more that one arrangement for the same model.
Keywords: Fixed-point iteration; Global solution methods; Occasionally binding constraints; Macroeconomic models
JEL Codes: E17; E44; F34; F41
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| FIPIT (F53) | reduction in execution time (C69) |
| FIPIT (F53) | comparable accuracy to traditional methods (C52) |
| FIPIT (F53) | ability to solve models without nonlinear solvers (C61) |