Finance: Market efficiency in the age of machine learning
Professor Leonidas Barbopoulos, University of Edinburgh
'Market Efficiency in the Age of Machine Learning'
Wednesday 18 May, 1pm - 2.15pm
As machines replace humans in financial markets, how is informational efficiency impacted? We shed light on this issue by exploiting a unique data-set that allows us to identify when machines access important company information (8-K filings) versus when humans access the same information. We find that increased information access by cloud computing services significantly improves informational efficiency and reduces the price drift following information events. We address identification through exogenous power and cloud outages, a quasi-natural experiment, and instrumental variables. We show that machines are better able to handle linguistically complex filings, less susceptible to bias from negative sentiment and less constrained in attention/processing capacity than humans.
Leonidas joined The University of Edinburgh as an Associate Professor (Reader) of Finance in July 2019. Prior to joining The University of Edinburgh, he was an Associate Professor of Finance at the Adam Smith Business School (University of Glasgow) and an Assistant Professor of Financial Economics within the School of Economics and Finance at the University of St. Andrews. Leonidas is a visiting Research Professor of Finance at NYU Stern School of Business (New York University, NY) and regularly visits as an Associate Professor of Finance the Mason School of Business at the College of William & Mary (VA) and the Harvard Business School at Harvard University (MA). Leonidas is a fellow of the UK Higher Education Academy and an Associate Editor at the International Review of Financial Analysis.
Further information: email@example.com
First published: 24 March 2022