Modeling Stock Order Flows and Learning Market-Making from Data

Technical Report CBCL Paper No. 217 / AI Memo No. 2002-009, M.I.T., Cambridge, MA

8 Pages Posted: 28 Jul 2008 Last revised: 30 Jul 2008

See all articles by Adlar J. Kim

Adlar J. Kim

Massachusetts Institute of Technology (MIT)

Christian R. Shelton

University of California, Riverside (UCR)

Tomaso Poggio

Massachusetts Institute of Technology (MIT) - Department of Brain and Cognitive Sciences

Date Written: June 1, 2002

Abstract

Stock markets employ specialized traders, market-makers, designed to provide liquidity and volume to the market by constantly supplying both supply and demand. In this paper, we demonstrate a novel method for modeling the market as a dynamic system and a reinforcement learning algorithm that learns profitable market-making strategies when run on this model.

The sequence of buys and sells for a particular stock, the order flow, we model as an Input-Output Hidden Markov Model fit to historical data. When combined with the dynamics of the order book, this creates a highly non-linear and difficult dynamic system. Our reinforcement learning algorithm, based on likelihood ratios, is run on this partially-observable environment. We demonstrate learning results for two separate real stocks.

Keywords: reinforcement learning, market making, order flow, market simulation

JEL Classification: C11, C13, C15, C32, C51, C52, C53

Suggested Citation

Kim, Adlar J. and Shelton, Christian R. and Poggio, Tomaso, Modeling Stock Order Flows and Learning Market-Making from Data (June 1, 2002). Technical Report CBCL Paper No. 217 / AI Memo No. 2002-009, M.I.T., Cambridge, MA, Available at SSRN: https://ssrn.com/abstract=1171370 or http://dx.doi.org/10.2139/ssrn.1171370

Adlar J. Kim (Contact Author)

Massachusetts Institute of Technology (MIT) ( email )

77 Massachusetts Avenue
Cambridge, MA 02139-4307
United States

Christian R. Shelton

University of California, Riverside (UCR) ( email )

900 University Avenue
Riverside, CA CA 92521
United States

Tomaso Poggio

Massachusetts Institute of Technology (MIT) - Department of Brain and Cognitive Sciences ( email )

Artificial Intelligence Labratory
Cambridge, MA 02139
United States
(617) 253-5230 (Phone)

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

Paper statistics

Downloads
798
Abstract Views
2,717
Rank
61,836
PlumX Metrics