MODELING CONSUMER PRICES FOR DRINKING MILK IN THE REGIONS OF THE RUSSIAN FEDERATION USING MACHINE LEARNING TECHNOLOGIES

Due to the large variability in average consumer prices for dairy products in the regions of the Russian Federation, an analysis of its causes and determining factors is relevant. The purpose of the work is to build a machine learning model that reflects the dependence of average consumer prices for drinking whole pasteurized milk with 2,5–3,2% fat content in the regions of the Russian Federation on indicators characterizing the production and consumption of dairy products, population size, and income. As a result, based on statistical data for 2016–2021. a model based on the gradient boosting algorithm was built. The significance of the model lies in the possibility of using it to identify and analyze the influence of individual regional characteristics (milk production, number of dairies, number of cows, milk yield, rural and urban population, per capita income, etc.) on average consumer prices for pasteurized milk in the regions RF. Understanding the factors in the formation of consumer prices on a regional basis is of practical importance for trading enterprises selling dairy products, as well as for dairy industry enterprises.