Прогнозирование динамики ВВП на основе данных мониторинга предприятий, проводимого Банком России

Abstract

The article is devoted to modeling and forecasting of the gross domestic product of Russia based on the data of enterprise survey conducted by the Bank of Russia. The purpose was to study the possibility of improving the efficiency and accuracy of GDP forecasting based on the use of these data. The authors used business climate indicators as well as balances of responses to questions about dynamics of output, demand, and prices. The methodology included regression analysis, as well as a combination of factor and regression analysis, both on aggregate data on all economic sectors and on data on individual economic branches. The forecasting results for two control samples containing actual observations for the eight quarters of 2017–2018 and the five quarters of 2022–2023 suggested that the use of monitoring data improved forecasting accuracy compared to the ARIMA benchmark model. The lowest estimation errors were obtained for approaches combining factor and regression analysis on aggregate data for all economic sectors or the economy as a whole.