Forecasting Asymmetries in Aggregate Stock Market Returns : Evidence from Conditional Skewness

This paper provides a time-series test for the Differences-of-Opinion theory proposed by Hong and Stein (2003) in the aggregate market, thus extending Chen, Hong, and Stein's (2001) cross-sectional test for this theory across individual stocks. An autoregressive conditional density model with a skewed-t distribution is used to estimate the effects of past trading volume on return asymmetry. Using NYSE and AMEX data from 1962 to 2000, we find that the prediction of the Hong-Stein model that negative skewness will be most pronounced under high trading volume conditions is not supported in our time-series analysis with market data