Cojumps in stock prices: empirical evidence

Dudley Gilder, Mark B. Shackleton, Stephen J. Taylor

Research output: Contribution to journalArticle

Abstract

We examine contemporaneous jumps (cojumps) among individual stocks and a proxy for the market portfolio. We show, through a Monte Carlo study, that using intraday jump tests and a coexceedance criterion to detect cojumps has a power similar to the cojump test proposed by Bollerslev et al. (2008). However, we also show that we should not expect to detect all common jumps comprising a cojump when using such coexceedance based detection methods. Empirically, we provide evidence of an association between jumps in the market portfolio and cojumps in the underlying stocks. Consistent with our Monte Carlo evidence, moderate numbers of stocks are often detected to be involved in these (systematic) cojumps. Importantly, the results suggest that market-level news is able to generate simultaneous large jumps in individual stocks. We also find evidence of an association between systematic cojumps and Federal Funds Target Rate announcements.
Original languageEnglish
Pages (from-to)443-459
Number of pages17
JournalJournal of Banking and Finance
Volume40
Issue number1
Early online date13 May 2013
DOIs
Publication statusPublished - 1 Mar 2014

    Fingerprint

Bibliographical note

© 2013, Elsevier. Licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International http://creativecommons.org/licenses/by-nc-nd/4.0/

Keywords

  • high-frequency stock prices
  • non-parametric jump tests
  • realised volatility
  • macroeconomic news

Cite this