Multifactor econometric export models of the Russian Federation region

The article is devoted to the results of econometric modeling of the dependence of the region’s exports on socio-economic factors according to the statistics of 2018. Statistics were collected for 84 regions of the Russian Federation based on data from the Federal State Statistics Service for 2018. The quality of the selected data on emissions was checked according to the Grubbs criterion, as a result of which emissions were removed from the statistical sampling - values for Moscow and St. Petersburg. A system of socio-economic indicators has been formed that affect the region’s exports, and significant factors have been identified: retail trade turnover, average annual number of employees, average monthly nominal accrued wages of employees of organizations, the volume of shipped goods of own production and services performed on our own, for mining and manufacturing, agricultural products, a balanced financial result in the economy, fixed assets, investment capital, per capita cash income. Various multivariate econometric models were estimated using the least squares method, their comparison and selection of the best in the sense of statistical quality was carried out. Gauss-Markov conditions verified. Heteroskedasticity was eliminated by the weighted least squares method. In the work, point and interval forecasts for regional exports are constructed. A comparison is made with the results of econometric modeling of regional exports according to 2017 data. The results can be used in the study of foreign economic activity and for its forecasting.