Analysis a short-term time series of crop sales based on machine learning methods

Abstract

The main goal of this article is to solve the problem associated with identifying sales seasons in time series in order to build the most accurate forecast of sales of various crops and provide decision support and improve the efficiency of business processes of agro-industrial companies. In this regard, the necessity of developing an algorithm that allows to form a time series of sales in accordance with the seasons available in it to improve the accuracy of existing sales forecasting methods is justified. This study provides a detailed description of the problem and its solutions in the form of an algorithm, as well as a comparison of the accuracy of building prediction models before and after its application, which confirms the consistency of the developed method for the formation of time series.