Detecting Iceberg Orders from Order Book Data
Summary
The article explains how limit order books work and why large visible orders can expose traders to front running and adverse selection. Iceberg orders reduce that exposure by displaying only part of a larger order, though hiding demand can also make execution slower. The article contrasts simple probing with statistical inference: a small limit order inside the spread may reveal hidden liquidity, while spread and displayed size relationships can be used to estimate iceberg likelihood from historical data.
Its more specific method identifies a venue’s refresh pattern: a trade, a reduction in displayed top-of-book size, and a replenishing order arrive close together. That sequence can indicate an iceberg and reveal its displayed clip size. A historical probability model then estimates total and remaining size from observed clip sizes. The article stresses that estimates remain probabilistic, the event pattern depends on exchange rules, and venue-specific data must guide model design. It presents the approach mainly as information for liquidity providers, not as a reliable way to trade against hidden orders.
Key ideas
- An order book records displayed bids and offers, and trades occur when orders can be matched.
- Iceberg orders conceal most of a large order to reduce the market impact of revealing full demand.
- Historical order book features can support probabilistic estimates of hidden liquidity, but correlations alone may be unreliable in live trading.
- A rapid sequence of a trade, displayed-size reduction, and order replenishment can reveal an iceberg on venues that use this refresh process.
- The detection pattern and size model depend on exchange-specific rules and should be grounded in observed data.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.