What is it about?
Machine Learning methods have been recently employed in Financial Economics as an alternative to other conventional techniques thanks to their flexibility in high-dimensional settings and their great prediction accuracy. I review and critically asses the most recent contributions in Asset Pricing, summarizing the empirical findings and providing hints forfuture developments.
Featured Image
Photo by Kevin Ku on Unsplash
Why is it important?
Machine Learning in Asset Pricing represents a very promising and dynamic field that has been growing steadily over the last years. A sound grasp of these methods is imperative for both practitioners and academics willing to keep abreast of the latest frontier of research. This work groups the most recent contrinbutions into 5 categories based on the main approach used, guiding the reader towards a sound economic interpretation of the findings, and points out the econometric challenges one must be wary of when applying Machine Learning in this area.
Perspectives
The goal of the article is to summarize the main econometric issues one encounters in Asset Pricing and to provide a concise overview of the results one can achieve in this field using non-standard methods known as Machine Learning. Focussing on the economic intrepretation of the methods employed, it aims at eradicating the skepticism that still remains towards them, showing that they are not just "black boxes" but rather they are quite intuitive after carefully examining their mechanics.
Matteo Bagnara
Leibniz Institute for Financial Research SAFE
Read the Original
This page is a summary of: Asset Pricing and Machine Learning: A critical review, Journal of Economic Surveys, September 2022, Wiley,
DOI: 10.1111/joes.12532.
You can read the full text:
Contributors
The following have contributed to this page
