Through the looking glass: Indirect inference via simple equilibria


Journal of Econometrics

April 2015, vol. 185, n°2, pp.343-358

Departments: Finance

Keywords: Hidden Markov model, Long-run risk, Learning, Value at risk, Indirect inference, Particle filters

This paper develops an indirect inference (Gourieroux et al., 1993; Smith, 1993) estimation method for a large class of dynamic equilibria. Our approach consists of constructing econometrically tractable auxiliary equilibria, obtained by simplifying the economic primitives of the structural equilibrium, via which estimation can proceed. We use this approach to develop an accurate estimator for the long-run risk model of Bansal and Yaron (2004). We demonstrate the method in Monte Carlo simulations and implement it on U.S. data. We also illustrate the good performance of the methodology on an equilibrium model with investor learning