Understanding Alpha Decay
研究了阿尔法衰减如何影响资产定价研究中已实现和条件预期收益的测量，发现阿尔法衰减会导致计量偏差，并提供了校正公式。
I study the importance of alpha decay for the measurement of realized and conditional expected returns in asset pricing studies. Alpha decay refers to the reduction in abnormal expected returns (relative to an asset pricing model) in response to an anomaly becoming widely known among market participants. As decreases in alpha are associated (ceteris paribus) with positive realized returns, the econometrician may misinterpret these repricing returns as evidence that the anomaly will persist in the future. Because alpha decay is generally a nonstationary phenomenon, asset pricing tests that impose stationarity may lead to biased inference. I illustrate the importance of alpha decay using the most commonly studied anomalies in the asset pricing literature and find that the measured alpha differs from the true alpha by about 1.4% per year. I provide a simple formula to correct for this bias and show how to incorporate alpha decay tests into the standard asset pricing toolkit. This paper was accepted by David Simchi-Levi, finance.