Deep Learning for Statistical Arbitrage in Similar Equities

Summary

The document presents a framework for statistical arbitrage that combines asset pricing, time-series modeling, and portfolio optimization. It first uses conditional latent pricing factors to identify similar assets and construct residual portfolios, then extracts trading signals with a convolutional transformer. Those signals feed a constrained policy designed to maximize risk-adjusted returns.

The empirical study uses daily US equity data and reports consistently strong out-of-sample mean returns and Sharpe ratios relative to benchmark approaches. The summary does not provide sample dates, transaction-cost treatment, risk details, or specific benchmark results, so the claims cannot be assessed fully from this description. The stated evidence concerns historical out-of-sample performance; it does not establish that returns will persist in live trading or after implementation costs.

Key ideas

• Residual portfolios based on conditional latent pricing factors are used to represent arbitrage opportunities among similar assets.
• A convolutional transformer extracts time-series signals from the residual portfolios.
• The signals determine a constrained trading policy targeting risk-adjusted returns.
• The reported evaluation uses daily US equities and describes out-of-sample performance against benchmarks.
• The brief description omits implementation details needed to judge real-world profitability.

Tags

# Deep Learning Statistical Arbitrage


# Deep Learning Statistical Arbitrage









Statistical arbitrage exploits temporal price differences between similar assets. We develop a unifying 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, that maximizes risk-adjusted returns under constraints. Our comprehensive empirical study on daily US equities shows a high compensation for arbitrageurs to enforce the law of one price. Our arbitrage strategies obtain consistently high out-of-sample mean returns and Sharpe ratios, and substantially outperform all benchmark approaches.

Shown in full with attribution under the source's licence. Licence: abstract CC0

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.