Forecasting Inflation in Russia Using Dynamic Model Averaging

Konstantin Styrin
Bank of Russia, New Economic School

Abstract

In this study, I forecast CPI inflation in Russia using the method of Dynamic Model Averaging pseudo out-of-sample on historical data. This method can be viewed as an extension of Bayesian Model Averaging, where the identity of the model that generates data is allowed to change over time, as are the model parameters. DMA is shown not to produce forecasts superior to simpler benchmarks, even if a subset of individual predictors is pre-selected ‘with the benefit of hindsight’ from the full sample. The two groups of predictors that give the highest average values for the posterior inclusion probability are loans to non-financial firms and individuals, along with actual and anticipated wages.