Short Term E-commerce Sales Forecast Method Based on Machine Learning Models

DOI

10.2991/978-2-494069-31-2_119How to use a DOI?

Keywords

Short-term forecast; Machine learning; Discount information

Abstract

Nowadays, e-commerce is developing rapidly in the world. In 2010, China’s e-commerce turnover reached 37.21 trillion yuan. For modern e-commerce corporations, an accurate sales forecast is the key to driving the development of corporations. While many effective forecast methods have been established in multiple business contexts, few of these methods have achieved good results in the short-term forecast and the value of detailed classified information of promotional plans has not yet been explored. This study attempts to establish a short-term forecast framework and explore whether incorporating detailed promotional plans can improve the forecast accuracy of the forecasting framework established. This study proposes a short-term forecast framework and implements six machine learning models to forecast daily sales. It finds that in a short-term forecast with one month’s data, the framework proposed can achieve rather good performance with out-of-sample MAPE ranging from 10.23% to 20.83% in different machine learning models. The incorporation of the detailed classification of discount information results in statistically significant improvements in the out-of-sample accuracy of linear regression, ridge regression, and lasso regression, with the best improvement of 36.19% in MAPE, but has no significant influence on the support vector machine, gradient boosting and random forest. From these results, the study provides recommendations for short-term forecast sales in general as well as a detailed classification of discount information.

Copyright

© 2022 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Tingli Feng
AU  - Chenming Niu
AU  - Yuchen Song
PY  - 2022
DA  - 2022/12/29
TI  - Short Term E-commerce Sales Forecast Method Based on Machine Learning Models
BT  - Proceedings of the 2022 6th International Seminar on Education, Management and Social Sciences (ISEMSS 2022)
PB  - Atlantis Press
SP  - 1020
EP  - 1030
SN  - 2352-5398
UR  - https://doi.org/10.2991/978-2-494069-31-2_119
DO  - 10.2991/978-2-494069-31-2_119
ID  - Feng2022
ER  -