Research Article
Sales Forecast of Walmart on Account of Multivariate Regression and Machine Learning Methods
@INPROCEEDINGS{10.4108/eai.28-10-2022.2328455, author={Zhaoyu Chen}, title={Sales Forecast of Walmart on Account of Multivariate Regression and Machine Learning Methods}, proceedings={Proceedings of the International Conference on Financial Innovation, FinTech and Information Technology, FFIT 2022, October 28-30, 2022, Shenzhen, China}, publisher={EAI}, proceedings_a={FFIT}, year={2023}, month={4}, keywords={sales prediction of walmart; multiple linear regression model; elastic-net regression model; polynomial regression model}, doi={10.4108/eai.28-10-2022.2328455} }- Zhaoyu Chen
Year: 2023
Sales Forecast of Walmart on Account of Multivariate Regression and Machine Learning Methods
FFIT
EAI
DOI: 10.4108/eai.28-10-2022.2328455
Abstract
Sales prediction is an integral part of modern commercial sales problems. Machine learning, especially supervised machine learning algorithms, can be used to find complex and unpredictable trends in dynamic sales problems, including various potentially influential variables. The success of each sales company is closely tied to accurate sales predictions, which can help enterprises find potential risks and make more sensible decisions. This paper aims to predict the future weekly sales of Walmart based on three different regression models (Multiple Linear Regression, Elastic-Net Regression, and Polynomial Regression). In addition, statistical metrics (e.g., R2 and RMSE) are adopted to evaluate the quality of the model. According to the analysis, relevant variables that had an obvious effect on weekly sales are the Holiday, Date, Type, and Stores. In terms of the prediction model, the simplest Multiple Linear Regression Model makes the best sales prediction, which has a moderate R2-Score of about 0.933 and the RMSE with the smallest difference between the training and test sets. Overall, these results shed light on guiding the selection of appropriate models in sales prediction as well as providing suggestions to retail companies to make sales strategies.