This paper surveys the burgeoning field of financial machine learning. It reviews the key concepts and arguments for the use of machine learning in finance, covering the latest applications in return prediction, factor models, stochastic discount factors, and portfolio choice. The authors explore the benefits and limitations of various machine learning methods and highlight the challenges of applying them to finance, especially when dealing with small datasets and noisy information. Additionally, the paper examines the economic implications of complexity, arguing that larger, more complex models often outperform simpler ones when dealing with misspecified data. The paper concludes by encouraging researchers to utilize machine learning tools to better understand economic mechanisms and to develop solutions to complex structural models.
Join Robinhood
https://join.robinhood.com/johnh1039
Paper Reference:
Financial Machine Learning
159 Pages Posted: 13 Jul 2023
Yale SOM; AQR Capital Management, LLC; National Bureau of Economic Research (NBER)
University of Chicago - Booth School of Business