Measuring Portfolio Gains: The Case of Earnings Announcement Trading Signals
研究提出新方法评估交易信号的实际收益，发现基于盈余公告漂移的信号能提高夏普比率，但收益集中在公告前后几天，且考虑市场摩擦后效果减弱。
ABSTRACT This study offers solutions to help address practical issues researchers generally overlook when assessing gains from a trading signal. Specifically, the methodologies used in previous research generally ignore investor risk aversion when forming portfolios, do not update portfolios as signals arrive, exploit look-ahead biases, do not assess the incremental gains of a new signal, and do not consider market frictions. We examine trading signals based on post-earnings announcement drift (PEAD), the earnings announcement premium (EAP), and earnings announcement rescheduling (RES). Using our proposed approach, we find that portfolios that incorporate the individual signals produce higher Sharpe ratios than equal-weighted portfolios; however, the gains for each signal are concentrated to a few days around the announcement. The EAP and RES signals do not provide incremental portfolio gains over the PEAD signal. After considering market frictions, portfolio performance rapidly attenuates and becomes similar to the SPY ETF as the portfolio size increases. JEL Classifications: G12; G14; G17.