Estimation of the discontinuous leverage effect: Evidence from the NASDAQ order book

TL;DR

This paper introduces robust nonparametric methods to detect and analyze the relationship between jumps in prices and volatility in high-frequency NASDAQ data, revealing conditional dependencies based on jump sign and type.

Contribution

It develops new methods for assessing the discontinuous leverage effect that are robust to market microstructure noise and applies them to six years of NASDAQ data.

Findings

No unconditional negative correlation between price and volatility cojumps.

Strong relation between cojumps when conditioned on jump sign and type.

Firm volatility levels explain cross-sectional variation in discontinuous leverage.

Abstract

An extensive empirical literature documents a generally negative correlation, named the "leverage effect," between asset returns and changes of volatility. It is more challenging to establish such a return-volatility relationship for jumps in high-frequency data. We propose new nonparametric methods to assess and test for a discontinuous leverage effect --- i.e. a relation between contemporaneous jumps in prices and volatility. The methods are robust to market microstructure noise and build on a newly developed price-jump localization and estimation procedure. Our empirical investigation of six years of transaction data from 320 NASDAQ firms displays no unconditional negative correlation between price and volatility cojumps. We show, however, that there is a strong relation between price-volatility cojumps if one conditions on the sign of price jumps and whether the price jumps are market-wide or idiosyncratic. Firms' volatility levels strongly explain the cross-section of discontinuous leverage while debt-to-equity ratios have no significant explanatory power.