Inference in Long-Horizon Event Studies: A Bayesian Approach with Application to Initial Public Offerings


  • Alon Brav

    Search for more papers by this author
    • Duke University, Fuqua School of Business. This paper has benefited from the comments of Brad Barber, Michael Barclay, Kobi Boudoukh, Michael Brandt, George Constantinides, Zvi Gilula, Paul Gompers, John Graham, David Hsieh, Shmuel Kandel, S. P. Kothari, Craig Mackinlay, Ernst Maug, Roni Michaely, Mark Mitchell, Nick Polson, Haim Reisman, Jay Ritter, Matthew Rothman, Jay Shanken, Erik Stafford, Robert Stambaugh, René Stulz (the editor), Richard Thaler, Sheridan Titman, Ingrid Tierens, Tuomo Vuolteenaho, Jerold Warner, Bob Whaley, Bob Winkler, two anonymous referees, and seminar participants at Boston College, Columbia, Cornell, Dartmouth, Duke, Harvard, London Business School, Ohio State, Rochester, Tel-Aviv, UCLA, Yale, and the 1999 AFA conference. I thank Jay Ritter for the IPO data set used in this study and Michael Bradley for access to the SDC database. Krishnamoorthy Narasimhan provided excellent research assistance. I owe special thanks to Eugene Fama, Campbell Harvey, and J. B. Heaton for their invaluable insights. All remaining errors are mine.


Statistical inference in long-horizon event studies has been hampered by the fact that abnormal returns are neither normally distributed nor independent. This study presents a new approach to inference that overcomes these difficulties and dominates other popular testing methods. I illustrate the use of the methodology by examining the long-horizon returns of initial public offerings (IPOs). I find that the Fama and French (1993) three-factor model is inconsistent with the observed long-horizon price performance of these IPOs, whereas a characteristic-based model cannot be rejected.