Capturing dynamics of post-earnings-announcement drift using genetic algorithm-optimised supervised learning

Ye and Schuller (2020) study event-driven strategies in equities using tree ensembles. The abstract mentions an out-of-sample test and the size of the universe but not a benchmark, transaction costs or multiple testing.

“First, we show how Post-Earnings-Announcement Drift can be analysed using machine learning methods and demonstrate such methods' prowess in producing credible forecasting on the drift direction.”

Also stated: Quarterly, Long-short, Fundamentals, out-of-sample test, universe of 1,106 companies

Missing from abstract: benchmark, costs, multiple testing