Dissecting cross-impact on stock markets: An empirical analysis

TL;DR

This paper introduces a multivariate linear propagator model to analyze cross-impact and interactions between stock markets, revealing significant contributions to return covariance and outperforming simplified models in out-of-sample tests.

Contribution

The paper presents a novel multivariate propagator model that captures sectorial correlations and cross-impact effects in stock markets, improving understanding of joint asset dynamics.

Findings

The model explains a significant portion of stock return covariance.

Simplified models outperform in out-of-sample predictions.

Market impact interactions are crucial for understanding asset correlations.

Abstract

The vast majority of market impact studies assess each product individually, and the interactions between the different order flows are disregarded. This strong approximation may lead to an underestimation of trading costs and possible contagion effects. Transactions in fact mediate a significant part of the correlation between different instruments. In turn, liquidity shares the sectorial structure of market correlations, which can be encoded as a set of eigenvalues and eigenvectors. We introduce a multivariate linear propagator model that successfully describes such a structure, and accounts for a significant fraction of the covariance of stock returns. We dissect the various dynamical mechanisms that contribute to the joint dynamics of assets. We also define two simplified models with substantially less parameters in order to reduce overfitting, and show that they have superior out-of-sample performance.