Analysis of Product Sales Volume and Pricing Based on Multivariate Linear Regression Model and Time Series Forecasting Algorithm

This paper presents an analysis of product sales and pricing based on a multi-linear regression model and time series forecasting algorithm. Initially, historical sales prices are considered cost-plus pricing. A multi-linear regression model is then established, with the monthly sales volume of each product category as the dependent variable and the cost-plus pricing of products within the category as the independent variable, to ascertain the relationship between sales volume, cost-plus pricing, and expected profit rates for each product category. Subsequently, utilizing data from the first seven days of July for the previous three years and considering the sales of individual products and historical wholesale prices, a time series forecasting algorithm is developed to predict the wholesale prices of products sold from July 1 to July 7, 2023. Finally, an objective programming model is constructed, with the expected profit rates as decision variables and the maximization of supermarket profits as the objective, to establish constraints and compute the pricing and sales volume for 31 saleable products. The maximum profit for the supermarket over 7 days is determined to be 12808.92568 RMB.