Testing Chinese Equity Order-Flow Factors for Stock Selection
Summary
The study groups fifty Wind order-flow measures into eight categories, including inflows, outflows, net active buying, and large-order flows. It evaluates these signals as stock-selection factors using portfolio sorts, cross-sectional regressions, and information coefficient and information ratio analysis. The tests examine factor separation, monotonicity, stability, and performance across company sizes and industries, while accounting for a stated one-way turnover cost assumption.
Across the examined period, four categories showed some stock-selection usefulness. The study highlights large-order outflow counts in portfolio sorts and the opening large-order net-inflow rate in regression and IC analysis. It also reports that order-flow measures retain incremental information after controlling for momentum, turnover, and volatility factors. These are historical results from a short holding horizon; the source warns that factor performance may change and that turnover can make costs substantial. Some reported statistics are missing in the supplied text, so they cannot be assessed here.
Key ideas
- Order-flow factors are built from order data and aim to capture micro-level supply and demand in equities.
- The study groups fifty measures into eight categories based on definitions and correlations.
- It evaluates factors with portfolio sorts, regression statistics, and IC and IR measures.
- Several flow categories showed historical stock-selection value, with results varying by test method.
- Residual analysis suggests the flow factors contain information beyond momentum, turnover, and volatility, subject to historical and trading-cost limitations.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.