Adaptive Optimization of Quantitative Strategies during the Macro Transformation Period from the Perspective of Machine Learning

DOI

10.2991/978-94-6463-835-6_71How to use a DOI?

Keywords

adaboost; quantitative finance; Adaptive optimization; feature selection; quantitative investment model

Abstract

Against the backdrop of global economic transformation characterized by post-pandemic supply chain realignments and geopolitical tensions, the financial market has experienced unprecedented volatility, challenging traditional quantitative investment frameworks’ foundational assumptions. This paper addresses the diminishing effectiveness of conventional factors—such as value and momentum—in low-interest-rate environments by integrating machine learning techniques. Through a hybrid methodology combining panel data analysis and ensemble modeling, this paper systematically evaluate the adaptability of Adaboost-based models across diverse market conditions. Specifically, this paper analyze the time-varying impact of macroeconomic variables ( such as policy rates and inflation expectations) on factor performance and propose a dynamic feature selection framework that incorporates technical indicators and liquidity metrics. This study employs a mixed-methods approach, including vector autoregressive (VAR) modeling to identify structural breaks in factor-return relationships and supervised learning to quantify the marginal contributions of new features. By analyzing 5-year daily data of ETFs tracking the CSI 300, CSI 500, CSI 1000, and S&P 500 indices, we document a 27-41% decline in traditional factor accuracy between 2022 and 2024. Crucially, our optimized model—incorporating RSI, Bollinger Bands, and VWAP alongside stepwise regression—achieves a 0.83 accuracy score, outperforming baseline models by 73% in out-of-sample testing.

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  - Zili Huang
PY  - 2025
DA  - 2025/09/17
TI  - Adaptive Optimization of Quantitative Strategies during the Macro Transformation Period from the Perspective of Machine Learning
BT  - Proceedings of the 2025 3rd International Academic Conference on Management Innovation and Economic Development (MIED 2025)
PB  - Atlantis Press
SP  - 666
EP  - 674
SN  - 2352-5428
UR  - https://doi.org/10.2991/978-94-6463-835-6_71
DO  - 10.2991/978-94-6463-835-6_71
ID  - Huang2025
ER  -