Confronting Machine Learning with Financial Research

Abstract

This article aims to examine the challenges and applications of machine learning for financial research. Machine learning algorithms have been developed for certain data environments that substantially differ from the one we encounter in finance. Not only do difficulties arise owing to some of the idiosyncrasies of financial markets, there is also a fundamental tension between the underlying paradigm of machine learning and the research philosophy in financial economics. Given the peculiar features of financial markets and the empirical framework within social science, various adjustments must be made to the conventional machine learning methodology. The authors discuss some of the main challenges of machine learning in finance and examine how these could be accounted for. Despite the challenges, the authors argue that machine learning could be unified with financial research to become a robust complement to the econometrician’s toolbox. Moreover, the authors discuss the various applications of machine learning in the research process, such as estimation, empirical discovery, testing, causal inference, and prediction.

TOPICS: Big data/machine learning, financial crises and financial market history, portfolio theory

Key Findings

• ▪ The authors compare machine learning to conventional quantitative research methodologies in finance.

• ▪ The authors discuss the idiosyncrasies of finance and the challenges that financial markets pose to machine learning methodologies.

• ▪ The authors examine the opportunities (and applications) that machine learning offers for financial research.