Abstract
I examine 468 estimates on the relationship between trading volume and stock returns reported in 44 studies. I deploy recent nonlinear techniques for detecting publication bias together with Bayesian and frequentist model averaging to evaluate the heterogeneity in the estimates.
The results yield three key conclusions. First, publication bias distorts the findings of the primary studies.
After this bias is corrected, the literature shows that with higher trading volume, returns decline in both effects in the contemporaneous and even in the dynamic one. Second, one cannot rely on any general conclusions about stock markets.
The predictability of stock returns varies with different markets and stock types. Third, different data characteristics, structural variations and methodologies used drive the heterogeneity in the results of the primary articles.
In particular, one should be cautious when using monthly data or VAR models.