BAYESIAN ESTIMATION OF REAL BUSINESS CYCLE MODEL: THE CASE OF POLAND

This paper presents the result of Bayesian estimation of real business cycle model augmented with mechanism of indivisible labor. We evaluate the structural parameters using quarterly data for Poland from 1997 to 2011. Methodology of the estimation is based on Bayes theorem. We use Kalman filter to estimate likelihood function and Metropolis algorithm to obtain posterior distribution for structural parameters. Our results are as follows. Firstly, we find that all estimated parameters are significantly greater than zero. We also find significantly greater value of the elasticity of output with respect to capital stock than it is assumed in standard calibration. The fit of model is relatively good taking under consideration model's simplicity. Moreover, the analysis of model's dynamics shows that capital stock and consumption are highly persistent in response to positive shock in technology.