The strategy clusters stocks based on momentum signals and firm characteristics using PCA and agglomerative clustering. It pairs stocks with the highest and lowest returns, going long on low-return stocks and short on high-return stocks.
ASSET CLASS: stocks | REGION: United States | FREQUENCY:
Monthly | MARKET: equities | KEYWORD: Machine Learning
I. STRATEGY IN A NUTSHELL
Use unsupervised ML clustering on 48 momentum and 78 firm characteristics to form stock pairs. Go long low-return, short high-return within each cluster. Equally weighted, rebalanced monthly.
II. ECONOMIC RATIONALE
Pairs trading exploits mean reversion: historically correlated stocks diverging creates profit opportunities. Clustering by firm characteristics improves pair selection, enhancing performance and robustness beyond classical statistical arbitrage.
III. SOURCE PAPER
Pairs Trading via Unsupervised Learning [Click to Open PDF]
Han, Chulwoo, Durham University; He, Zhaodong, Durham University; Toh, Alenson Jun Wei, Nanyang Technological University
<Abstract>
This paper develops a pairs trading strategy via unsupervised learning. Unlike conventional pairs trading strategies that identify pairs based on return time series, we identify pairs by incorporating firm characteristics as well as price information. Firm characteristics are revealed to provide important information for pair identification and significantly improve the performance of the pairs trading strategy. Applied to the US stock market from January 1980 to December 2020, the long-short portfolio constructed via the agglomerative clustering earns a statistically significant annualized mean return of 24.8% and a Sharpe ratio of 2.69. The strategy remains profitable after accounting for transaction costs and removing stocks below 20% NYSE-size quantile. A host of robustness tests confirm that the results are not driven by data snooping.
IV. BACKTEST PERFORMANCE
| Annualised Return | 24.8% |
| Volatility | 9.2% |
| Beta | N/A |
| Sharpe Ratio | 2.69 |
| Sortino Ratio | N/A |
| Maximum Drawdown | -12.3% |
| Win Rate | N/A |
