Abstract
We investigate the specification and power of intraday event study test statistics. Mean, market, and matched firm models generate well-specified return results for a range of intervals up to 60 min around the event. These models detect return shocks equivalent to one spread in one-minute interval data and three spreads in longer intervals. Researchers using intraday return event studies can, therefore, be confident in their robustness. Some volume event study approaches have reasonable power but they are not generally well specified, while a matched-firm approach gives the best combination of specification and power for spread event studies.
Acknowledgments
We thank Andrea Bennett, Adrian Lee, participants at the 2018 New Zealand Finance Colloquium, and the Research Group at Foreign Trade University, Vietnam for comments that have improved the paper. We are also very grateful to our editor, Guochang Zhang, and an anonymous referee for valuable comments. All errors are our own.
Notes
1 MacKinlay (Citation1997) suggests event studies data back as far as Dolley (Citation1933).
2 See Binder (Citation1998) and Corrado (Citation2011) for excellent reviews.
3 See for instance, Kothari and Warner (Citation1997), Fama (Citation1998), Lyon et al. (Citation1999), Brav (Citation2000), Mitchell and Stafford (Citation2000), Kothari and Warner (Citation2004), Jegadeesh and Karceski (Citation2009), and Viswanathan and Wei (Citation2008).
4 We thank an anonymous referee for suggesting this analysis.
5 Volume is measured as the natural logarithm of raw volume. See Bamber, Barron, and Stevens (Citation2011) for an excellent discussion around volume measurement.