Time Variations and Covariations in the Expectation and Volatility of Stock Market Returns
实证研究股票收益的条件均值与波动率的协动关系，发现商业票据与国债利差可预测波动率变化，且波动率领先预期收益，质疑将预期收益建模为条件波动率常函数的做法。
ABSTRACT This article investigates empirically the comovements of the conditional mean and volatility of stock returns. It extends the results in the literature by demonstrating the role of the commercial paper—Treasury yield spread in predicting time variation in volatility. The conditional mean and volatility exhibit an asymmetric relation, which contrasts with the contemporaneous relation that has been tested previously. The volatility leads the expected return, and this time series relation is documented using offset correlations, short‐horizon contemporaneous correlations, and a vector autoregression. These results bring into question the value of modeling expected returns as a constant function of conditional volatility.