Statistical Arbitrage in Rank Space

TL;DR

This paper shows that representing stocks by their rank in market capitalization reveals better mean-reversion properties, enabling more effective statistical arbitrage strategies than traditional name-based approaches, especially when using neural networks.

Contribution

It introduces a novel rank space representation for market dynamics and demonstrates its superiority for statistical arbitrage, incorporating neural networks and intraday rebalancing.

Findings

Rank space representation outperforms name space in statistical arbitrage.

Neural networks enhance portfolio performance in rank space.

Intraday rebalancing improves conversion between rank and name space portfolios.

Abstract

Equity market dynamics are conventionally investigated in name space where stocks are indexed by company names. In contrast, by indexing stocks based on their ranks in capitalization, we gain a different perspective of market dynamics in rank space. Here, we demonstrate the superior performance of statistical arbitrage in rank space over name space, driven by a robust market representation and enhanced mean-reverting properties of residual returns in rank space. Our statistical arbitrage algorithm features an intraday rebalancing mechanism for effective conversion between portfolios in name and rank space. We explore statistical arbitrage with and without neural networks in both name and rank space and show that the portfolios obtained in rank space with neural networks significantly outperform those in name space.