Optimal Liquidation Through a Limit Order Book: A Neural Network and Simulation Approach

Alexandre Roch ()
Additional contact information
Alexandre Roch: University of Quebec at Montreal

Methodology and Computing in Applied Probability, 2023, vol. 25, issue 1, 1-29

Abstract: Abstract We present a learning algorithm based on simulation and neural networks to solve a stochastic optimal control problem with a large state space using dynamic programming. The problem consists in liquidating a given number of shares of a stock through a limit order book (LOB). The state space includes prices and quantities at each level in the LOB. The objective is to maximize the expected liquidation proceeds. Shares are sold by market orders matching the current limit orders in the LOB and have an impact on the future evolution of the LOB. The optimal strategy is obtained using a hybrid form of performance and value iteration procedures based on neural networks. The probability distribution of the LOB is estimated by a deep learning classification task. The model is tested on 12 stocks traded through the NYSE Arcabook, and a numerical implementation shows that the model outperforms the most common optimal liquidation models in the literature by a significant amount.

Keywords: Stochastic control problem; Optimal liquidation; Limit order books; Neural networks; 91G60; 93E20 (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
http://link.springer.com/10.1007/s11009-023-09996-z Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:metcap:v:25:y:2023:i:1:d:10.1007_s11009-023-09996-z

Ordering information: This journal article can be ordered from
https://www.springer.com/journal/11009

DOI: 10.1007/s11009-023-09996-z

Access Statistics for this article

Methodology and Computing in Applied Probability is currently edited by Joseph Glaz

More articles in Methodology and Computing in Applied Probability from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().