When Quotes Crumble: Detecting Transient Mechanical Liquidity Erosion in Limit Order Books

TL;DR

This paper presents a neural network-based framework for detecting transient liquidity erosion in limit order books, validated using a simulated multi-agent environment with ground truth data.

Contribution

It introduces a novel detection pipeline and neural model that reliably identify crumbling quotes, outperforming rule-based methods across various market conditions.

Findings

Neural model achieves +36% AUC over rule-based baselines.

Framework performs robustly across different market regimes.

Simulation provides ground truth for transient liquidity events.

Abstract

We study the detection of transient liquidity erosion ("crumbling quotes") in electronic limit order books, where observable quote deterioration may reflect either mechanical liquidity withdrawal or informational repricing. Using the ABIDES agent-based simulator, we construct a multi-agent environment in which crumbling emerges from stochastic regime switches in a market maker, providing time-resolved ground truth unavailable in real market data. We develop a detection pipeline that identifies mechanically driven quote erosion using order book features, and train a neural model to produce calibrated crumbling probabilities. Experiments demonstrate that the proposed framework reliably identifies crumbling events against agent-level ground truth, with the neural model achieving +36% AUC improvement over rule-based baselines and robust performance across normal, high-volatility, bull, and bear market conditions. Ablation studies on temporal features and varying the dependence structure of the ground-truth mechanism confirm that the framework generalizes across both independent and autocorrelated liquidity withdrawal dynamics.