Opacity and event study analysis : a thesis presented in partial fulfillment of the requirements for the degree of Master of Finance at Massey University, Auckland, New Zealand

Abstract

This study examines the impact of asset pricing model’s misspecification on the power of an event study analysis. Gilbert, Hrdlicka, Kalodimos, and Siegel (2014) show that asset pricing model fails to price asset accurately at high frequency. This is due to uncertainty about the effect of systematic news on firm value, which they address as firm opacity. They propose an additional factor in the market model and empirically show better performance of the augmented model. This study practically investigates the implication of this additional factor on enhanced power of event study analysis. Key findings indicate that an adjusted asset pricing model improves the power of event studies for small stock portfolio. The detection rate increases from 2.9 percent to 15.5 percent based on an induced abnormal return of 1.5 percent to 2 percent. However, there is no improvement in abnormal return detectability in portfolios of random stocks or other characteristic- sorted portfolios.