Determination of the main factors of the gross regional product level by econometric modeling methods for the totality of regions of the Russian Federation

Abstract

The purpose of the research is to identify factors that determine the level of gross regional product, using the construction of econometric models. In terms of the economic theory the volume of gross added value of the region depends on existence of resources, efficiency of use of resources, level of effective demand for the consumption of gross added value, level of export, structure of the regional economy. In the article, by constructing econometric models with a different set of factors, the most essential and statistically significant factors determining the level of gross regional product were identified.Materials and methods. Official data of Federal State Statistics Service for 2016 formed empirical base of the research. The aggregate of the constituent entities of the Russian Federation was considered as a statistical aggregate as an object of study, and both the complete aggregate of subjects and the aggregate of entities with a gross regional product level of less than and more than a trillion rubles were studied. Research method is econometric modeling of gross regional product based on spatial data on constituent entities of the Russian Federation. The parameters of the multiple linear regression models in absolute values and in logarithms were estimated using the least squares method. To test the null hypotheses, a five percent significance level was adopted. Interval estimation of model parameters was carried out taking into account standard errors consistent under heteroskedasticity. The quality of models was defined by coefficient of determination, mean relative error of approximation, Ramsey’s test checking a null hypothesis about lack of the passed variables in model. Implementation of the research was carried out in the RStudio software product.Results. Eight qualitative models of multiple linear regression are constructed. Two models are constructed on the general set of constituent entities of the Russian Federation: 1) model in absolute values with three quantitative signs and one attributive sign characterizing group of entities on the level of a gross regional product; 2) three-factor model in logarithms. Two models are constructed on set of entities with a gross regional product less than one trillion rubles: 1) model in absolute values with seven quantitative signs, five of which were statistically significant; 2) four-factor model in logarithms. Four two-factor models are constructed on set of entities with a gross regional product more than one trillion rubles.Conclusion. In all constructed models the most significant factors of the level of gross regional product are: a) factor of fixed capital (average annual residual cost of fixed assets); b) factor of effective demand (expressed by one of the following indicators: social expenses of the budget, total income of the population, wage fund); c) factor of level of financial resources (expressed by the volume of average annual balances on ruble accounts of legal persons and balance of incomes and expenses of the enterprises and organizations). At creation of econometric models statistically insignificant were factors of resource efficiency (capital productivity, labor productivity, profitability), factors characterizing the volume of foreign economic activity and a share of the extractive industries. The economy of the regions in 2016 can be described as an economy of an extensive type, focusing on the use of domestic capital and financial resources.