Machine Learning and Expected Returns

59 Pages Posted: 2 Feb 2023

See all articles by Julio A. Crego

Julio A. Crego

Tilburg University

Jens Soerlie Kvaerner

Tilburg University

Marc Stam

Tilburg University

Date Written: February 2, 2023


Financial data is characterized by a low signal-to-noise ratio making it difficult to identify robust functional forms that map the characteristics of financial securities to expected returns (Lettau and Pelger, 2020). In this paper, we modify the standard prediction problem in empirical asset pricing by replacing realized returns with an estimator for expected return developed by Martin and Wagner (2019). We use a neural network to map expected returns to 164 stock characteristics and their interactions with eight macroeconomic time-series, resulting in 1476 predictors. Portfolios based on the predictions from the neural network generate risk-adjusted returns with respect to the Fama-French 6-factor model in the range of 1.4% (t-statistic of 3.04) to 1.2% (t-statistic of 2.65) before and after transaction costs; out-of-sample. The corresponding Sharpe ratios are 1.15 and 1.06. A similar analysis based on realized returns results in Sharpe ratios below the market portfolio.

Keywords: Machine Learning, Big Data, Equity Options, Return Prediction, Cross-Section of Returns, Transaction Costs

JEL Classification: G12, G11

Suggested Citation

Crego, Julio and Soerlie Kvaerner, Jens and Stam, Marc, Machine Learning and Expected Returns (February 2, 2023). Available at SSRN: or

Julio Crego

Tilburg University ( email )

P.O. Box 90153
Tilburg, DC Noord-Brabant 5000 LE

Jens Soerlie Kvaerner (Contact Author)

Tilburg University ( email )

Warandelaan 2
Tilburg, -- 4818HK
40242704 (Phone)
0364 (Fax)


Marc Stam

Tilburg University ( email )

P.O. Box 90153
Tilburg, 5000 LE
0642112812 (Phone)

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Abstract Views
PlumX Metrics