Review of Financial Studies Vol. 36 No. 6 2023

Man versus Machine Learning: The Term Structure of Earnings Expectations and Conditional Biases

JULES H. Van BINSBERGEN¹; Xiao Han²; Alejandro Lopez-Lira³

¹ The Wharton School, University of Pennsylvania , NBER, and CEPR, USA · ² Bayes Business School, City, University of London , UK · ³ Warrington College of Business, University of Florida , USA

Abstract

We introduce a real-time measure of conditional biases to firms’ earnings forecasts. The measure is defined as the difference between analysts’ expectations and a statistically optimal unbiased machine-learning benchmark. Analysts’ conditional expectations are, on average, biased upward, a bias that increases in the forecast horizon. These biases are associated with negative cross-sectional return predictability, and the short legs of many anomalies contain firms with excessively optimistic earnings forecasts. Further, managers of companies with the greatest upward-biased earnings forecasts are more likely to issue stocks. Commonly used linear earnings models do not work out-of-sample and are inferior to those analysts provide.

DOI

10.1093/rfs/hhac085

Volume

36

Issue

6

Pages

2361-2396

Language

en

Sources

bibtex:phds-export.bib openalex crossref

Cite

APA 7 MLA 9 Chicago (author–date) Harvard Elsevier Harvard (JFE, JBF…) The Accounting Review Academy of Management American Marketing Association INFORMS (Management Science) Copy citation Copied ✓