Using Limit Order Book Imbalance to Study Short-Term Price Changes

Summary

This article explains two signals derived from limit order books: volume imbalance at the best bid and ask, and order flow imbalance based on changes in displayed size and quote prices. The first compares the quantities resting at the best buy and sell prices, with values near either extreme indicating a stronger side. The second tracks additions and removals at those quotes, aiming to emphasize recent activity over older resting orders. It also describes extending the calculation across several price levels and testing signals with regression models.

The evidence summarized includes a chart relating first-level imbalance to subsequent price changes and an analysis of commodity order-book data that associates buying pressure with more positive short-horizon moves, and selling pressure with more negative ones. The article cautions that the average predicted changes remain within the bid-ask spread, so predictive association alone does not establish a profitable trading strategy. Its findings are summaries of cited research, not a complete implementation or transaction-cost-adjusted evaluation.

Key ideas

• Best bid and ask sizes can be combined into a normalized measure of displayed buying versus selling pressure.
• Order flow imbalance tracks changes at the best quotes to focus on more recent activity.
• The article describes extending imbalance measures across multiple levels of the book.
• The cited evidence associates imbalance with short-horizon price moves, but average moves remain within the spread.
• Signal prediction does not by itself demonstrate profitability after spreads and execution costs.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.