Market Stability with Machine Learning Agents

25 Pages Posted: 17 May 2019 Last revised: 17 Jan 2020

See all articles by Christophre Georges

Christophre Georges

Hamilton College - Department of Economics

Javier Pereira

Hamilton College - Department of Economics

Date Written: April 18, 2019

Abstract

We consider the effect of adaptive model selection and regularization by agents on price volatility and market stability in a simple agent-based model of a financial market. The agents base their trading behavior on forecasts of future returns, which they update adaptively and asynchronously through a process of model selection, estimation, and prediction. The addition of model selection and regularization methods to the traders' learning algorithm is shown to reduce but not eliminate overfitting and resulting excess volatility. Our results suggest that even a high degree of attention to overfitting on the part of traders who are engaged in data mining is unlikely to entirely eliminate destabilizing speculation. They also accord well with the empirical ``sparse signals" and ``pockets of predictability" findings of Chinco, Clark-Joseph and Ye (2019) and Farmer, Schmidt and Timmermann (2019).

Keywords: Expectations, Agent-Based Modeling, Machine Learning, LASSO, Asset Prices, Volatility

JEL Classification: D83 D84 D53, E44, G14 G17

Suggested Citation

Georges, Christophre and Pereira, Javier, Market Stability with Machine Learning Agents (April 18, 2019). Available at SSRN: https://ssrn.com/abstract=3374666 or http://dx.doi.org/10.2139/ssrn.3374666

Christophre Georges (Contact Author)

Hamilton College - Department of Economics ( email )

198 College Hill Road
Clinton, NY 13323
United States
315 859-4472 (Phone)
315 859-4477 (Fax)

HOME PAGE: http://people.hamilton.edu/christophre-georges

Javier Pereira

Hamilton College - Department of Economics ( email )

198 College Hill Road
Clinton, NY 13323
United States

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
65
Abstract Views
402
rank
358,614
PlumX Metrics