We apply state-of-the-art financial machine learning to assess the return-predictive value of more than 45,000 earnings … announcements on a majority of S&P1500 constituents. To represent the diverse information content of earnings announcements, we … generate predictor variables based on various sources such as analyst forecasts, earnings press releases and analyst conference …

We selectively review recent advancements in research on predictive models of earnings and returns. We discuss why … applying statistical, econometric, and machine learning advancements to forecasting earnings and returns presents difficult …