Deep Learning Statistical Arbitrage
Jorge Guijarro-Ordonez, Markus Pelger, Greg Zanotti
Management Science · Thu
Abstract
Statistical arbitrage exploits temporal price differences between similar assets. We develop a comprehensive conceptual framework for statistical arbitrage and a novel data-driven solution. First, we construct arbitrage portfolios of similar assets as residual portfolios from conditional latent asset pricing factors. Second, we extract their time-series signals with a powerful machine learning time-series solution, a convolutional transformer. Lastly, we use these signals to form an optimal trading policy, which maximizes risk-adjusted returns under constraints. Our comprehensive empirical study on daily U.S. equities shows a high compensation for arbitrageurs to enforce the law of one price. Our arbitrage strategies obtain considerable out-of-sample mean returns and Sharpe ratios, and outperform all benchmark approaches. This paper was accepted by Agostino Capponi, finance. Funding: The authors thank MSCI for generous research support.
Keywords
statistical arbitrage, pairs trading, machine learning, deep learning, big data, stock returns, convolutional neural network, transformer, attention, factor model, market efficiency, investment
Enable JavaScript for citation tools and additional paper interactions.