Review of Economic Approaches and Models for GDP Forecasting

Abstract

This study systematizes the known forecasting methods applicable to the assessment of future GDP dynamics in both long- and short-term prediction periods. Aim. This study aims to compare economic approaches and models used to forecast the macroeconomic indicators of a country’s socioeconomic development, including GDP. Tasks. The study analyzes GDP dynamics in Russia, reviews economic approaches and models used in forecasting, summarizes the applied methods for forecasting indicators of the system of national accounts, determines groups of forecasting methods, and provides an expert assessment of GDP for 2015 and statistical forecasting based on a multivariate nonlinear regression model. Methods. This study uses general methods of scientific cognition and comparative and retrospective analyses to systematize econometric, expert, and combined approaches applicable to GDP forecasting in the context of dynamic socioeconomic and geopolitical transformations. Results. Known forecasting methods applicable to the pre-emptive assessment of GDP and other macro- and microeconomic indicators of socioeconomic development are examined. Statistical data describing GDP dynamics in the Russian Federation are summarized. The authors provide their results of GDP forecasting according to the Cobb-Douglas production function. Conclusions. It is substantiated that three groups of methods can be applied in GDP forecasting - factographic (statistical), expert (intuitive), and combined - and these groups differ in the nature of the data required for forecasting. The authors conclude that the practical application of a particular GDP forecasting method is determined by the availability of required information, selected forecasting method, competence of the forecaster, purpose of the forecast, and length of the prediction period.