Theoretically Motivated Data Augmentation and Regularization for Portfolio Construction
This work addresses a theoretical gap for quantitative finance practitioners, though it is incremental as it builds on existing empirical deep learning methods.
The authors tackled the lack of theoretical understanding for data augmentation in deep learning for portfolio construction, showing that injecting noise scaled by the square root of past returns improves performance over no noise or irrelevant techniques.
The task we consider is portfolio construction in a speculative market, a fundamental problem in modern finance. While various empirical works now exist to explore deep learning in finance, the theory side is almost non-existent. In this work, we focus on developing a theoretical framework for understanding the use of data augmentation for deep-learning-based approaches to quantitative finance. The proposed theory clarifies the role and necessity of data augmentation for finance; moreover, our theory implies that a simple algorithm of injecting a random noise of strength $\sqrt{|r_{t-1}|}$ to the observed return $r_{t}$ is better than not injecting any noise and a few other financially irrelevant data augmentation techniques.