AlphaForge: A Framework to Mine and Dynamically Combine Formulaic Alpha Factors
Hao Shi, Weili Song, Xinting Zhang, Jiahe Shi, Cuicui Luo, Xiang Ao, Hamid Arian, Luis Seco
TL;DR
AlphaForge presents a two-stage framework for quantitative alpha discovery and dynamic combination: a generative-predictive mining model creates high-fitness, diverse formulaic alphas, while a dynamic factor-timing model selects and weights factors day-by-day to form Mega-Alpha. By replacing fixed-factor weights with time-adaptive weights and maintaining interpretability, the approach addresses non-stationarity in markets and factor obsolescence. Empirical results on CSI300/CSI500 demonstrate superior IC, RankIC, and portfolio performance, with real-money trading showing profitable excess returns. The work advances formulaic alpha research by integrating gradient-based factor generation with a flexible, transparent combination mechanism suitable for live investment settings.
Abstract
The complexity of financial data, characterized by its variability and low signal-to-noise ratio, necessitates advanced methods in quantitative investment that prioritize both performance and interpretability.Transitioning from early manual extraction to genetic programming, the most advanced approach in the alpha factor mining domain currently employs reinforcement learning to mine a set of combination factors with fixed weights. However, the performance of resultant alpha factors exhibits inconsistency, and the inflexibility of fixed factor weights proves insufficient in adapting to the dynamic nature of financial markets. To address this issue, this paper proposes a two-stage formulaic alpha generating framework AlphaForge, for alpha factor mining and factor combination. This framework employs a generative-predictive neural network to generate factors, leveraging the robust spatial exploration capabilities inherent in deep learning while concurrently preserving diversity. The combination model within the framework incorporates the temporal performance of factors for selection and dynamically adjusts the weights assigned to each component alpha factor. Experiments conducted on real-world datasets demonstrate that our proposed model outperforms contemporary benchmarks in formulaic alpha factor mining. Furthermore, our model exhibits a notable enhancement in portfolio returns within the realm of quantitative investment and real money investment.
