Using Order Book Imbalance for Intraday Signals and Monthly Stock Selection

Summary

This study examines whether differences in buy and sell limit order strength contain information about short-term price moves and longer-horizon stock returns in Chinese equities. Its Spread measure aggregates order book strength at each tick. Across prediction windows from 5 to 200 ticks, relatively thicker bids are associated with a greater chance of an upward short-term move. At daily frequency, however, thick displayed bids are associated with stronger realized aggressive selling and weaker cross-sectional returns, so the measure’s interpretation changes with sampling horizon.

The authors convert high-frequency observations into a monthly stock-selection factor and report backtests spanning 2010 through August 2019. After neutralization, the long-short portfolio’s annualized return was 14.6%, with an information ratio of 2.01; the factor also showed reported predictive information across several Chinese equity universes and limited overlap with common price-volume measures. These are historical backtest results, not proof of future returns. The document flags factor, model, and market-style changes as risks.

Key ideas

• Order book bid and ask strength can provide information about near-term price direction.
• The Spread measure’s interpretation differs between tick-level and daily analysis.
• A monthly factor derived from high-frequency data showed stock-selection ability in the reported backtest.
• The reported factor added information relative to several common price-volume measures.
• Historical performance may not persist if the factor, model, or market style changes.

Tags

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