Time-varying analysis of dynamic stochastic general equilibrium models based on sequential Monte Carlo methods

摘要

A new method is proposed to estimate parameters, natural rates, and unknown states of dynamic stochastic general equilibrium models simultaneously, based on the particle filter and a self-organizing state space model. We estimate the parameters and the natural rates using the time-varying-parameter approach, which is often used to infer invariant parameters practically. In most previous works on DSGE models, structural parameters of them are assumed to be deep (invariant). However, our method analyzes how stable structural parameters are. Adopting the TVP approach creates the great advantage that the structural changes of parameters are detected naturally. Moreover, we estimate time-varying natural rates of macroe-conomic data: real output, inflation rate, and real interest rate. The fit of a DSGE model is evaluated using the log-likelihood of it. Thus, we are able to compare the fits of DSGE models. In empirical analysis, we estimate a new Keynesian DSGE model using the US data.