Research and Analysis on XGBoost, Random Forest, and Their Hybrid Models in Retail Sales Forecasting

DOI

10.2991/978-94-6239-801-6_138How to use a DOI?

Keywords

Retail Sales Forecasting; XGBoost; Random Forest; Hybrid Model; Machine Learning

Abstract

Retail sales forecasting plays a crucial role in inventory management, marketing planning, and supply-chain optimization, and the choice of model directly influences how well retailers respond to demand volatility, promotions, and seasonal patterns. With the growing complexity of retail sales forecasting, the use of machine learning models for data analysis has become increasingly important. This study reviews relevant literature to summarize the performance of XGBoost and Random Forest in retail sales forecasting, and analyzes the effectiveness of hybrid models combining these two algorithms. XGBoost features high prediction accuracy, effective regularization to mitigate overfitting, and strong capability in modeling nonlinear relationships and imbalanced data. Random Forest demonstrates high stability, robustness to noise, and excellence in feature importance analysis. Hybrid models combining XGBoost and Random Forest achieve more accurate and reliable forecasting results than single models. This study provides practical guidance for model selection in retail sales forecasting. In future work, more competitive hybrid models will be further explored and evaluated.

Copyright

© 2026 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  - Zedong Wang
PY  - 2026
DA  - 2026/10/23
TI  - Research and Analysis on XGBoost, Random Forest, and Their Hybrid Models in Retail Sales Forecasting
BT  - Proceedings of the 2026 2nd International Conference on Financial Innovation and Marketing Management (FIMM 2026)
PB  - Atlantis Press
SP  - 1331
EP  - 1337
SN  - 2352-5428
UR  - https://doi.org/10.2991/978-94-6239-801-6_138
DO  - 10.2991/978-94-6239-801-6_138
ID  - Wang2026
ER  -