Monolith Research
In plain terms
Stocks tend to keep drifting in the direction of an earnings surprise for weeks after the announcement — the post-earnings-announcement drift. This studies whether that classic effect is still harvestable at scale.
How it works
Earnings surprises are ranked across a broad point-in-time universe, holding a diversified basket of the largest positive surprises against the largest negative ones.
What it’s tested against
Run on a survivorship-correct panel with earnings actuals and estimates, costs stressed, to separate a real residual drift from one already arbitraged away.
Data
IBES estimates and actuals; a CRSP / Compustat point-in-time panel.
Researched — evaluated at scale against the modern, more-efficient tape.