I reexamine the results of Hong et al. (2007).
Few industries can predict the market using extended period and data.
The results of Hong et al. are less significant after data revisions.
I find some evidence of the opposite predictive direction from the market to industries.
The overall results are consistent with market efficiency.