We use machine learning methods and high-dimensional detailed financial data to predict the direction of one-year-ahead earnings changes. Our models show significant out-of-sample predictive power: the area under the receiver operating characteristics curve ranges from 67.52% to 68.66%,...

The most prevalent forecasts of firms' long-term earnings issued by analysts are two-year-ahead earnings per share (EPS) estimates. When introduced by analysts, two-year-ahead EPS estimates set market expectations for firms' future earnings. Subsequent revisions to these estimates are highly...