Nowcasting Russian GDP in a Mixed-Frequency DSGE Model with a Panel of Non-Modelled Variables

Abstract

The study focuses on improving the accuracy of nowcasting Russian gross domestic product (GDP) growth rates using dynamic stochastic general equilibrium (DSGE) models. I modify one of the DSGE models of the Russian economy to incorporate mixed-frequency data by introducing an equation that links a panel of non-modelled high-frequency indicators of the current state of the economy to observable variables, whose dynamics are determined directly by the model. The results of the out-of-sample pseudo-real-time forecasting demonstrate that the incorporation of these additional variables enhances the accuracy of Russian GDP nowcasting using the DSGE model and makes the nowcast from this model comparable with forecasts from competing non-structural models and outperforms benchmark models in accuracy. The study also investigates the extent to which fluctuations in highfrequency indicators are associated with macroeconomic factors, as well as the economic shocks driving the explained portion of these fluctuations. I note that the structural interpretation of non-modelled variables is a potential strength of the model, though caution is warranted due to the econometric methodology employed.