Market Impact in Trader-Agents: Adding Multi-Level Order-Flow Imbalance-Sensitivity to Automated Trading Systems

TL;DR

This paper enhances automated trading agents by integrating multi-level order-flow imbalance sensitivity, leading to more realistic market impact effects in simulated markets, and critiques previous imbalance measures for robustness.

Contribution

It introduces a robust multi-level order-flow imbalance measure for trader-agents, improving their market impact simulation over prior methods.

Findings

Imbalance-sensitive trader-agents exhibit market impact effects.

The new measure outperforms previous imbalance metrics.

Enhanced agents are suitable for impact-aware trading environments.

Abstract

Financial markets populated by human traders often exhibit "market impact", where the traders' quote-prices move in the direction of anticipated change, before any transaction has taken place, as an immediate reaction to the arrival of a large (i.e., "block") buy or sell order in the market: e.g., traders in the market know that a block buy order will push the price up, and so they immediately adjust their quote-prices upwards. Most major financial markets now involve many "robot traders", autonomous adaptive software agents, rather than humans. This paper explores how to give such trader-agents a reliable anticipatory sensitivity to block orders, such that markets populated entirely by robot traders also show market-impact effects. In a 2019 publication Church & Cliff presented initial results from a simple deterministic robot trader, ISHV, which exhibits this market impact effect via monitoring a metric of imbalance between supply and demand in the market. The novel contributions of our paper are: (a) we critique the methods used by Church & Cliff, revealing them to be weak, and argue that a more robust measure of imbalance is required; (b) we argue for the use of multi-level order-flow imbalance (MLOFI: Xu et al., 2019) as a better basis for imbalance-sensitive robot trader-agents; and (c) we demonstrate the use of the more robust MLOFI measure in extending ISHV, and also the well-known AA and ZIP trading-agent algorithms (which have both been previously shown to consistently outperform human traders). We demonstrate that the new imbalance-sensitive trader-agents introduced here do exhibit market impact effects, and hence are better-suited to operating in markets where impact is a factor of concern or interest, but do not suffer the weaknesses of the methods used by Church & Cliff. The source-code for our work reported here is freely available on GitHub.