Estimating Nonlinear Business Cycle Mechanisms with Linear Vector Autoregressions: A Monte Carlo Study*

The paper investigates how well linear vector autoregressions (VARs) identify endogenous cycle mechanisms and frequencies when the underlying process is a nonlinear limit cycle. We conduct Monte Carlo simulations with five models in which cycles are driven by interaction of two state variables. find that while VARs quantitatively underestimate strength mechanism, they successfully qualitative presence mechanism most cases (55%–100%). Our results further suggest surprisingly successful at estimating processes.