Dynamic stochastic general equilibrium (DSGE) models have attracted the attention of policy makers as well as macroeconomists. They have a certain advantage over a vector autoregressive (VAR) model in the sense that DSGE models can identify various shocks in a theoretically consistent way since they are based on micro-foundation theories. As an estimation method for DSGE models, Bayesian analysis using MCMC techniques has recently become popular. Bayesian analysis also enables us to estimate a VAR model with the priors selected using a DSGE model, which is called DSGE-VAR model. This presentation explains the Bayesian analysis of DSGE and DSGE-VAR Models with its application to Japanese economy.