Forecasting with DSGE models with financial frictions

To investigate to what extent adding financial frictions can contribute to an improvement in the quality of DSGE model-based forecasts DSGE models with and without financial frictions. Comparison of point and density forecasts. The main finding is that accounting for financial frictions affecting firms tends to improve the quality of point forecasts while the opposite is true for the extension with household sector financial frictions. However, for all models point forecasts can be considered poor in the absolute sense and density forecasts are rather badly calibrated. We show that the main source of these problems is a significant and sizable bias in the forecasts for most of standard macroeconomic variables.