In this paper we compare two types of models for forecasting Russia’s GDP under the structural breaks. We consider models that allow breaks in a deterministic trend, in which the dates of structural breaks set exogenously, and more flexible class of models – with a stochastic trend. We show that models with a stochastic trend demonstrate the best result in GDP growth rates forecasting for a year ahead. For shorter horizons, the best forecasting model is the error correction model with a break in the deterministic trend in the GDP level.
Are you one of the authors of this document?