On-Line Portfolio Selection with Moving Average Reversion

TL;DR

This paper introduces the Moving Average Reversion (MAR) concept and the OLMAR strategy, which improve online portfolio selection by addressing limitations of single-period mean reversion assumptions, showing better performance and speed.

Contribution

It proposes a novel multiple-period mean reversion model and an online learning-based strategy, OLMAR, enhancing robustness and efficiency in portfolio selection.

Findings

OLMAR outperforms existing mean reversion algorithms on challenging datasets.

OLMAR achieves significantly better trading results.

OLMAR is computationally fast and practically applicable.

Abstract

On-line portfolio selection has attracted increasing interests in machine learning and AI communities recently. Empirical evidences show that stock's high and low prices are temporary and stock price relatives are likely to follow the mean reversion phenomenon. While the existing mean reversion strategies are shown to achieve good empirical performance on many real datasets, they often make the single-period mean reversion assumption, which is not always satisfied in some real datasets, leading to poor performance when the assumption does not hold. To overcome the limitation, this article proposes a multiple-period mean reversion, or so-called Moving Average Reversion (MAR), and a new on-line portfolio selection strategy named "On-Line Moving Average Reversion" (OLMAR), which exploits MAR by applying powerful online learning techniques. From our empirical results, we found that OLMAR can overcome the drawback of existing mean reversion algorithms and achieve significantly better results, especially on the datasets where the existing mean reversion algorithms failed. In addition to superior trading performance, OLMAR also runs extremely fast, further supporting its practical applicability to a wide range of applications.