隔夜報酬研究：於異常報酬之應用

Abstract

不可諱言，過去多數學者在探討報酬率時，並未將盤中和盤後之情況予以分離。然而，我們由Lou, Polk, and Skouras (2016)的研究，得以發現報酬率於盤中和盤後的表現，有莫大的差異。有鑑於此，我們決定將研究觸角伸入至過去文獻鮮少提及的盤後部分，深入探索隔夜報酬的表現。本研究的前半部分，主要是依循Lou et al. (2016)的研究，將報酬率拆解為盤中及盤後兩部分。本研究的後續發展，主要是依循Aboody, Even-Tov, Lehavy, and Trueman (Forthcoming 2017)的研究，將拆解而得的隔夜報酬視為情緒指標，探討相關議題。本研究顯示，相較於其他異常報酬，動能、獨特性波動度、獲利性較易受到投資人情緒的影響。除此之外，隔夜報酬能夠強化Keloharju, Linnainmaa, and Nyberg (2016)所述稱的報酬季節性效果(Return Seasonalities)，顯示投資人情緒在季節性策略中扮演重要的角色。我們認為，後續研究可望從情緒觀點出發，深入探究投資人情緒與季節性效果之實質關聯，並揭露其他可能影響季節性效果之系統因子。

We decompose the close-to-close returns into their intraday and overnight components on a great variety of popular trading strategies and firm characteristics, including price momentum, earnings momentum, industry momentum, profitability, investment, beta, idiosyncratic volatility, equity issuance, discretionary accruals, turnover, and one-month return. Particularly, we discover that most of the abnormal profits during the intraday and overnight periods have the opposite sign and both of them are statistically significant, indicating that decomposing returns into their intraday and overnight parts may be meaningful. Further, our study explores the application of overnight return in return anomalies. In particular, we regard overnight return as a measure of firm-specific investor sentiment and put forward four main findings. First, our results suggest that the long-leg portfolios of most anomaly strategies have a higher average return following high sentiment. Second, we find that half of the anomaly strategies have no significant relation between returns on the short leg and lagged sentiment. Third, we show that mispricing is more likely during high-sentiment periods than during low-sentiment periods in the anomalies such as momentum, profitability, and idiosyncratic volatility. Last but not least, we demonstrate that firm-specific investor sentiment can strengthen seasonality profits in most cases, indicating investor sentiment may be one of the shared systematic factors, which may shed light on return seasonalities.