ESTIMASI PENJUALAN PRODUK MENGGUNAKAN REGRESI LINEAR SEDERHANA PADA TOKO RITEL

Abstract

Retail sales transaction data represent a valuable source of information that can support decision-making, particularly in sales estimation and inventory management. This study aims to apply the Simple Linear Regression method to estimate product sales and identify the product category with the highest total sales. The research employed a quantitative approach using secondary data consisting of 1,000 retail sales transaction records. The research stages included data preprocessing, variable construction, the implementation of the Simple Linear Regression method, model evaluation using the Mean Absolute Percentage Error (MAPE), and product category analysis. The results produced the regression equation Y = 42,320.60 − 1,045.80X, with an estimated sales value of 27,679.40 for the 14th period. The model evaluation yielded a MAPE value of 153.78%, indicating that the estimation error remained relatively high. In addition, the Electronics category recorded the highest total sales of 156,905, making it a useful reference for inventory management. The findings indicate that the Simple Linear Regression method can be applied to estimate sales based on historical data; however, improving model accuracy requires more representative historical data or the application of alternative estimation methods.