Title: On posterior inferences of misspecified DSGE models: The minimal econometric interpretation
Authors: Takashi Kano - Hitotsubashi University (Japan) [presenting]
Abstract: The minimal econometric interpretation (MEI) has been previously developed for formal model evaluation and comparison among misspecified DSGE models in accompany by a theoretical reference model. The MEI approach recognizes a DSGE model as an incomplete econometric tool that provides only prior distributions of targeted population moments, but has no implication for actual data and sample moments. A reference model generates posterior distributions of targeted population moments conditional on data. The MEI approach, however, is absent from posterior updating of structural parameters of DSGE models. The conventional MEI approach is extended by proposing a simple method of drawing posterior inferences of DSGE models. The method exploits a Dirichlet-multinomial model to provide a mixture of prior and posterior distributions of targeted population moments implied respectively by DSGE and reference models. This simulation-based population moment-matching method is applied to draw posterior inferences of nonlinear structural asset pricing models.