Deniz Erdemlioglu, IÉSEG School of Management
Project Description/Abstract
We develop a new framework to measure systemic tail risk embedded in a panel of high-frequency stock returns. We estimate time-varying jump intensities and introduce test statistics that are conditional on the release times of news events. Our approach pinpoints when individual stocks or portfolio indices jump together at the high-frequency trading scales. Controlling for the multiple testing bias, we establish the consistency of the tests with bootstrap and show that the tests have reasonable finite sample performance. Our empirical analysis provides strong evidence that Federal Open Market Committee (FOMC) news creates systemic downside risk in stocks and sector-specific ETFs. We construct a simple proxy for news-driven systemic tail risk from the test statistics. This tail risk indicator helps explain the pre-FOMC announcement drift and reveals stock return predictability ahead of the upcoming Fed meeting. There is, however, no evidence that macro news creates systemic cojumps or crashes. We discuss the practical implications of our results for portfolio diversification and (news-driven) realized tail risk monitoring.
Co-authors
Christopher J. Neely, Federal Reserve Bank of St. Louis
Xiye Yang, Rutgers University