A0190
Title: Finite-state Markov-chain approximations: A hidden Markov approach
Authors: Eva Janssens - University of Amsterdam (Netherlands) [presenting]
Sean McCrary - University of Pennsylvania (United States)
Abstract: A novel finite-state Markov chain approximation method is proposed for Markov processes with continuous support. The method can be used for both uni- and multivariate processes, as well as non-stationary processes such as those with a life-cycle component. In contrast to existing methods, our discretization procedure provides both an optimal grid and transition probability matrix. We provide guidance on how to select the optimal number of grid points. The method is based on minimizing the information loss between a misspecified approximating model and the true data-generating process. The method outperforms existing discretization methods in several dimensions, including parsimoniousness. Furthermore, we demonstrate the performance of our discretization method compared to existing methods through the lens of an asset-pricing model and a life-cycle consumption-savings model. We find the choice of discretization method matters for the accuracy of the model solutions and for the welfare effects of risk.
