Can machine learning be helpful to asset management – and if so, how? Asset markets fundamentally differ from many of the environments in which machine learning has enjoyed success, and research into machine learning for asset management is just beginning.
In a thought piece from AQR, the firm’s Portfolio Solutions Group discusses the crucial points for understanding the current state of machine learning in the practice of asset management:
- How machine learning attempts to solve difficult problems and how it differs from traditional computer programming.
- Why finance poses unique challenges even for the most powerful machine learning programs
- Early research evidence hints that machine learning tools can potentially improve investment portfolios
The application of machine learning techniques is a natural evolution for investment research, and one that will continue to be explored.
Financial machine learning has the potential to be the next leap forward in quantitative investing. Two key points are crucial for understanding the current state of machine learning in the practice of asset management. The first is that research is just taking off and many important questions are yet to be answered. The second is that early research evidence indicates potential economically and statistically significant improvements in the performance of portfolios that leverage machine learning tools (Gu, Kelly, and Xiu 2018). However, the gains are evolutionary, not revolutionary.
The ideas behind machine learning—leveraging new data sets to identify robust additive portfolio performance and using quantitative methods to extract information systematically—are the modus operandi of quantitative investment processes. For decades, asset managers have used human-intensive, decentralized statistical learning. Machine learning offers a systematic approach to investing that mechanizes that process, allows managers to metabolize information from more new sources faster, including unstructured data previously untapped, and provides tools to search through increasingly flexible economic models that seek to better capture complex realities of financial markets. The evolution of machine learning in finance is just beginning.