Analysis of the regression model of the enterprise’s financial activity by research on residual error

Abstract

Purpose. Improvement of regression economic-mathematical models taking into account the influence of residual error as a random variable. Methodology. Methods of economic-mathematical modeling, regression analysis are used. The real conditional law of distribution of residual error as a complete characteristic of a random variable is applied. Findings. A scientific and practical approach to economic and mathematical modeling based on the study on residual error, to improve the construction of regression equations. Originality. For the first time, the application of residual error analysis as a random variable has been proposed in order to construct its conditional differential distribution function, which allows improving the quality of economic-mathematical modeling in the form of regression equations. The use of the proposed method of taking into account the residual error allows eliminating the negative impact of the violation of the conditions of the properties of the residual error in the implementation of economic and mathematical modeling using regression equations. Practical value. The analysis of the obtained results of economic-mathematical modeling of economic activity of Inhulets Mining and Processing Plant on significant statistical material with the use of the developed algorithm of residual error research confirmed the effectiveness of the proposed approach. It is recommended to include the developed algorithm taking into account the properties of the residual error in the practice of managing the financial activities of mining enterprises.