DSGE model with nominal rigidities: estimation and assessing of fit*

This paper deals with estimation of DSGE model of open economy with nominal rigidities and assessing of its data fit. The model is estimated using Bayesian techniques on data of Czech economy. Estimated values of structural parameters are economically interpreted. Assessing of model fit to data is carried out along several dimensions. Firstly, the fitted series are compared with observed series. Secondly, the volatility and the autocorrelations implied by the model are compared with the statistics calculated from the data. And thirdly, estimated unobserved shocks give picture of important events in the Czech economy. The results of the analysis show that model has problem in replicating import inflation and hours worked. Besides visual assessment, calculated standard deviations of model variables and observed variables indicate that volatility of wage inflation is higher in the model and volatility of import inflation is lower in the model than in the data. The auto correlations of nominal interest rate, CPI inflation and output growth are not matched well by the model. Estimates of unobserved shocks show that period of recession in 1997 had many sources. High nominal interest rates can be ascribed to monetary policy and risk premium shocks, decline in output growth is consequence of negative productivity shock. High inflation in late 90's is partly explained as rise in markups of domestic producers. Despite few failures, overall fit of the model to the data can be considered as satisfactory.