Hybrid Ridge–XGBoost Model with Advanced Feature Engineering for Sales Forecasting

DOI

10.2991/978-2-38476-497-6_33How to use a DOI?

Keywords

Retail Demand Forecasting; Time Series Prediction; Hybrid Machine Learning; Feature Engineering; Ridge–XGBoost

Abstract

Accurate sales forecasting is vital for business planning, inventory control, and promotions. While Ridge regression and XGBoost have been widely applied separately, this study proposes a novel hybrid combining their strengths in linear regularization and nonlinear boosted trees. Using the Rossmann Store Sales dataset (over one million transactions, 1,115 stores), we applied systematic preprocessing, advanced feature engineering (lags, rolling statistics, expanding means, cyclical and leave-one-out encoding), and Optuna-based hyperparameter tuning with a time-based split. The hybrid was evaluated in three stages: baseline (R² = 0.53), tuned (R² = 0.79), and engineered+tuned (R² = 0.99). Results show the proposed Ridge–XGBoost framework significantly outperforms individual models and benchmarks, offering a scalable and interpretable solution for retail demand forecasting.

Copyright

© 2025 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  - M. D. Monir Bhuyan
AU  - Xiaoyang Liu
PY  - 2025
DA  - 2025/12/15
TI  - Hybrid Ridge–XGBoost Model with Advanced Feature Engineering for Sales Forecasting
BT  - Proceedings of the 2025 International Conference on Educational Innovation and Information Technology (EIIT 2025)
PB  - Atlantis Press
SP  - 332
EP  - 338
SN  - 2352-5398
UR  - https://doi.org/10.2991/978-2-38476-497-6_33
DO  - 10.2991/978-2-38476-497-6_33
ID  - Bhuyan2025
ER  -