Using Regression Algorithms to Forecast Merchandise Sales in the Presence of Independent Variables

This article shows how forecasting retail merchandising volume can be done using machine learning approaches. In terms of research significance, the amount of merchandise sold is essential to business operations. Accurate forecasting can lower the amount of capital required by the company and increase the area for capital turnover. This study compares different supervised learning models for regression algorithms, including decision tree regression, random forest regression, ridge regerssion, and extra tree regression. Utilizing product sales, the data set was forecasted. Forecasting prduct sales is important in general because it is the largest retailer in the world. This paper's main contribution is to examine whether independent factors, such as weather, holidays, and fuel prices, might alter the algorithm of choice for best performance when predicting the volume of item sales at various venues.