Machine Learning based Stock Price Prediction: A Case Study of LVMH Group

DOI

10.2991/978-94-6239-801-6_145How to use a DOI?

Keywords

Stock Price Prediction; Machine Learning; Overfitting; LVMH Group

Abstract

In the current market, the high noise and non-stationary characteristics of stock price time series make accurate prediction extremely challenging. This study focuses on LVMH Group (MC.PA), the most representative blue chip asset in Europe, and extracts its real transaction data from January 2020 to May 2026 to construct lag features. The aim is to explore the predictive performance of machine learning models of different complexities in real financial noise environments. This study deployed both basic linear regression and complex ensemble tree models - Random Forest and Extreme Gradient Boosting (XGBoost). The empirical results reveal a counterintuitive core finding that, under the current lag characteristics, the simplest linear model achieved the lowest root mean square error (RMSE=10.3330) and the best fit (R²=0.9779) in the test set, significantly better than the RMSE of the random forest and XGBoost models (11.1151 and 11.1007, respectively). This result strongly demonstrates the almost random walk nature of financial asset prices, as well as the structural flaw of ensemble tree models being prone to overfitting in high noise data.

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  - Yu Wang
PY  - 2026
DA  - 2026/10/23
TI  - Machine Learning based Stock Price Prediction: A Case Study of LVMH Group
BT  - Proceedings of the 2026 2nd International Conference on Financial Innovation and Marketing Management (FIMM 2026)
PB  - Atlantis Press
SP  - 1396
EP  - 1402
SN  - 2352-5428
UR  - https://doi.org/10.2991/978-94-6239-801-6_145
DO  - 10.2991/978-94-6239-801-6_145
ID  - Wang2026
ER  -