微笑曲線與成交量預測股票未來報酬

Abstract

本論文探討買權和賣權各別成交量的資訊意涵，使用買權對股票成交量和賣權對股票成交量檢驗是否資訊領先者偏好使用買權和賣權兩者來交易負面資訊。對於隱含波動度的斜率(skew)，除了微笑曲線左邊的斜率外，我們也檢驗右邊的斜率，得到對股市未來報酬有預測性。由於反應在微笑曲線和成交量的資訊不盡相同，因此我們使用雙變數排序得到的投資組合對股票未來報酬有經濟上和統計上顯著的預測性。由於不只存在對股票價格方向的資訊，也存在對於股票波動度做交易的資訊，在建立投資組合時，可以除去兩個方向skew都很大的樣本，改善我們的投資組合報酬。對於交易策略，我們也檢驗了不同期間的表現，以及在不同的市場情況下觀察績效之差異。

We examine the information content of the call to stock volume ratio (CS) and the put to stock volume ratio (PS) to check whether informed traders prefer to trade bad news by trading both call and put options. Aside from the skew of the left-hand side of volatility smirk, we also examine the skew of the right-hand side of volatility smirk. Due to different information contents between option volume and skew measures, we apply double-sorted strategies by subsequently sorting option volume measures and option skew measures, which has a better cross-sectional predictive power for future stock returns than single-sorted strategies. Furthermore, because of the existence of informed trading for future stock volatility, we improve our double-sorting strategies by removing those samples with maximal ranks of both direction skew measures. Finally, we perform robustness analyses in different sub-periods and different market conditions.