Co-Trading Networks for Equity Covariance and Portfolio Allocation

Summary

The study uses the timing proximity of trades across US stocks to define co-trading relationships and build dynamic networks. It applies spectral clustering to these networks to identify groups of stocks whose dependencies evolve over time, complementing classifications based on industry sectors.

Using high-frequency limit order book data covering 2017 to 2019, the authors find a statistically significant positive relationship between low-latency co-trading and return covariance. They also develop a high-dimensional covariance estimator informed by the networks. Mean-variance portfolios using those estimates are reported to have lower volatility and higher Sharpe ratios than standard benchmarks. The excerpt does not specify benchmark construction, transaction costs, or robustness beyond the studied data period, so the portfolio findings require those details for practical assessment.

Key ideas

• Trade timing across stocks is used to measure co-trading and construct dynamic equity networks.
• Spectral clustering identifies economically meaningful stock groups whose dependencies change over time.
• The study reports a positive relationship between low-latency co-trading and return covariance.
• A network-informed covariance estimator supports mean-variance portfolios reported to outperform standard benchmarks on volatility and Sharpe ratio.

Tags

# 2302.09382


# Co-trading networks for modeling dynamic interdependency structures and estimating high-dimensional covariances in US equity markets









The time proximity of trades across stocks reveals interesting topological structures of the equity market in the United States. In this article, we investigate how such concurrent cross-stock trading behaviors, which we denote as co-trading, shape the market structures and affect stock price co-movements. By leveraging a co-trading-based pairwise similarity measure, we propose a novel method to construct dynamic networks of stocks. Our empirical studies employ high-frequency limit order book data from 2017-01-03 to 2019-12-09. By applying spectral clustering on co-trading networks, we uncover economically meaningful clusters of stocks. Beyond the static Global Industry Classification Standard (GICS) sectors, our data-driven clusters capture the time evolution of the dependency among stocks. Furthermore, we demonstrate statistically significant positive relations between low-latency co-trading and return covariance. With the aid of co-trading networks, we develop a robust estimator for high-dimensional covariance matrix, which yields superior economic value on portfolio allocation. The mean-variance portfolios based on our covariance estimates achieve both lower volatility and higher Sharpe ratios than standard benchmarks.

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.