Reinforcement Learning for Trade Execution with Market Impact

Patrick Cheridito and Moritz Weiss Department of Mathematics, ETH Zurich, Switzerland · 2025

Read Reinforcement Learning for Trade Execution with Market Impact by Patrick Cheridito and Moritz Weiss Department of Mathematics, ETH Zurich, Switzerland (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-06345.txt.

Description

Patrick Cheridito and Moritz Weiss apply reinforcement learning to trade execution in a 2025 paper. A trader who must buy or sell a large position wants the best total price. Each order placed in the book can move the market, so timing and order type both matter.

Their system learns how to split trades between market orders, which execute at once, and limit orders, which wait for a set price. A flexible probability model guides the split and keeps training efficient. In simulated order books with noise traders, tactical traders, and a rival strategic trader, the learned policy beat classic benchmark strategies on revenue. The work shows learning agents can handle live order books.

Catalog record

CREATORPatrick Cheridito and Moritz Weiss Department of Mathematics, ETH Zurich, Switzerland
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE60 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2507-06345

Browse the full library · About the Tower of Babel