КОРРЕЛЯЦИОННО-РЕГРЕССИОННЫЙ АНАЛИЗ РЫНКА АВТОМОБИЛЬНЫХ БЕНЗИНОВ

Abstract

Purpose of the study: to identify the factors affecting the volume of production of motor gasoline. Research methods: analysis and synthesis, systematic approach, as well as methods of correlation and regression analysis. Results of the research: A methodological approach to the use of tools for correlation and regression analysis of the gasoline market is proposed, which includes the following stages: 1) formation of a data array; 2) carrying out correlation analysis, building a correlation matrix, selecting factors into the model using the Correlation tool in the Data Analysis package of MS Excel; 3) conducting a regression analysis, constructing a regression equation, substantiating the obtained dependence using the "Regression" tool in the "Data Analysis" MS Excel package, calculating the elasticity coefficients. It is proposed to use the volume of production of motor gasoline as effective in carrying out the correlation-regression analysis and constructing mathematical models. Among the dependent variables and factors affecting the volume of production of motor gasoline, it is proposed to use variables that characterize four groups of factors: resource (raw material) limitations (the volume of oil production and refining), production capabilities of the industry (through the depth of oil refining and the yield of light oil products that characterize production capacity and set of installations in the industry), price attractiveness of the market (prices of producers and consumers of motor gasoline, world oil prices), export attractiveness (volume and value of exports). Multivariate economic and statistical models of the dependence of the volume of production of motor gasoline on a number of factors have been developed. Based on the results of calculations, a strong correlation was revealed between the volume of production of motor gasoline and the values of primary oil refining, oil production, and export of motor gasoline. The predicted values are located as close as possible to the residual values, which indicates that the resulting regression equation has a high degree of accuracy. Research prospects: the research results can be used to identify significant factors in the development of the motor gasoline market in the Russian Federation.