Meta-Learning the Optimal Mixture of Strategies for Online Portfolio Selection

TL;DR

This paper introduces a meta-learning based online portfolio selection model that dynamically combines multiple strategies to adapt to changing markets, emphasizing quick adaptation, transferability, and efficiency in high-frequency trading.

Contribution

It proposes a novel mixture policies framework with meta-learning for rapid adaptation and robust transferability in non-stationary financial environments.

Findings

Outperforms existing methods in training time and data efficiency.

Demonstrates strong transferability across different datasets.

Excels in high-frequency algorithmic trading scenarios.

Abstract

This paper presents an innovative online portfolio selection model, situated within a meta-learning framework, that leverages a mixture policies strategy. The core idea is to simulate a fund that employs multiple fund managers, each skilled in handling different market environments, and dynamically allocate our funding to these fund managers for investment. To address the non-stationary nature of financial markets, we divide the long-term process into multiple short-term processes to adapt to changing environments. We use a clustering method to identify a set of historically high-performing policies, characterized by low similarity, as candidate policies. Additionally, we employ a meta-learning method to search for initial parameters that can quickly adapt to upcoming target investment tasks, effectively providing a set of well-suited initial strategies. Subsequently, we update the initial parameters using the target tasks and determine the optimal mixture weights for these candidate policies. Empirical tests show that our algorithm excels in terms of training time and data requirements, making it particularly suitable for high-frequency algorithmic trading. To validate the effectiveness of our method, we conduct numerical tests on cross-training datasets, demonstrating its excellent transferability and robustness.