Anomalies within an Anomaly: A Five Factor Model Designed for the Understanding and Exploitation of the Post Earnings Announcement Drift

Post-earnings announcement drift (PEAD) is a well-documented anomalous stock price behavior. However, this anomaly does not occur around 36% of the time according to Zhipeng and Yan (2008). This study attempts to reconcile these "anomalies within an anomaly" within an information uncertainty framework. To address the multifaceted nature of information uncertainty causality, this study uses a five factor proxy for market uncertainty comprising measures of earnings momentum; firm size; liquidity; market sentiment; and earnings surprise. This study has three objectives. The first is to determine whether the unconventional initial response is limited in time and duration. The second is to directly relate the five-factor market uncertainty proxy to the elapsed time between the earnings announcement and price momentum reversal, and the magnitude of the price reversal. The third is to design a trading strategy, based on the five-factor market uncertainty proxy, to exploit the temporary nature of these "anomalies within an anomaly." The results of the study are mixed. The unconventional occurrences do return to a conventional trajectory. Although the five factors do not result in an uncertainty measure that is capable of definitively measuring or predicting the duration and/or magnitude of the anomalous occurrences' first response; the trends witnessed in the factors and when combining the factors into different scenarios do indicate possible directions to explore. These observations serve as the basis for the business rules in a heuristic-based trading strategy. The application and back testing of this trading strategy does outperform the benchmarks.