Bezirgen Veliyev (Aarhus University)

Title: Lights out: Jump risk after dark

Abstract: 

We show that companies' earnings releases constitute an important source of idiosyncratic jump risk in equity markets.  We construct a standardized measure of unexpected earnings per share (EPS) from analyst expectations and find that the size of the post-announcement price jump is highly correlated with the earnings surprise. As most companies release their results during market closure, we base our investigation on a data set that includes high-frequency information about transactions executed outside of the official exchange trading hours. We develop a jump test suitable for such relatively illiquid markets by extending the classical bipower variation-based jump test Barndorff-Nielsen and Shephard (2006) to a pre-averaged version, which can be implemented on noisy high-frequency data. A supplemental simulation study shows that our proposed test has good size control under a variety of noise structures, while it has excellent power under the alternative.