Estimating Permanent Price Impact via Machine Learning

93 Pages Posted: 3 Dec 2019

See all articles by Richard Philip

Richard Philip

University of Sydney Business School

Date Written: October 2, 2019


In this paper, we show that vector auto-regression (VAR) models, which are commonly used to estimate permanent price impact, are misspecified and can produce conflicting and incorrect inferences when the price impact function is nonlinear. We propose an alternative method to estimate permanent price impact by modifying a reinforcement learning (RL) framework. Our approach assumes the data is stationary and Markov, but is otherwise unrestrictive. We obtain empirical estimates for our model using an iterative learning rule and demonstrate that our model captures nonlinearities and makes correct inferences.

Keywords: Price impact; Information content of a trade; Machine learning; Reinforcement learning

JEL Classification: C45, C58, G14

Suggested Citation

Philip, Richard, Estimating Permanent Price Impact via Machine Learning (October 2, 2019). Journal of Econometrics, Forthcoming, Available at SSRN: or

Richard Philip (Contact Author)

University of Sydney Business School ( email )

Cnr. of Codrington and Rose Streets
Sydney, NSW 2006

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Abstract Views
PlumX Metrics