A Contrarian Trading Strategy: Mean vs. Median Reversion

Abstract

To optimize the total cumulative wealth, it needs to rebalance the portfolio on a period-by-period basis, using previously published portfolio values. The median and mean reversion techniques are two contrarian trading strategies. Mean reversion is a common trading strategy in portfolio theory. If it is used properly, it could outperform a benchmark. Using a strong -median estimator, median reversion explicitly predicts the next price vector. However, current mean and median reversion methods have several limitations: First, the mean reversion method does not consider noise and outliers; It suffers from estimate mistakes, resulting in suboptimal portfolios and poor performance. Second, although a median reversion method works well in the presence of noise and outliers, it also suffers the same issues when the dataset contains worthless data. As a result, the two methods may not work well on real-world datasets, which may contain both noise and worthless data. To address the issues mentioned above, we provide two methods for selecting an online portfolio. Also, we propose a hybrid reversion approach with a weighted scheme. To evaluate the effectiveness of our method, extensive experiments have been conducted on real-world datasets. It can improve performance and reduce the impact of noise and outliers.