Sensitivity Analysis of DSGE Model of Open Economy with Nominal Rigidities

This paper deals with sensitivity analysis of DSGE model estimated on data of Czech economy using Bayesian techniques. The main goal is comparison of different competing models in terms how they fit the data,. The benchmark model includes four types of nominal rigidities which are based on Calvo price setting mechanism. The competing models assume flexibility in some of these prices or combinations of them. The measure of fit depends on marginal likelihood obtained from Bayesian estimation. The. emphasis is also put on the sensitivity of estimated parameters for different specifications of the model. The results of estimation show that frictions in import prices and wages play the most significant role in the model of Czech economy. Sticky wages are far more important than sticky prices. The most sensitive parameters are Frisch elasticity of labor supply and openness of the economy. On the other hand, the measurement errors are quite stable across the models and have negligible effect on the results.