Testing Order-Flow Factors for Chinese Equity Selection

Summary

This study organizes fifty Wind order-flow measures into eight groups, including inflows, outflows, net active buying, and large-order net flows. It assesses their stock-selection value with industry-neutral portfolio sorts, cross-sectional regressions, and information coefficients, using ten-trading-day holding windows. The tests also compare the measures with turnover, momentum, and volatility factors and examine residuals to identify incremental information.

Four groups show some usefulness in the reported sample: inflow, outflow, large-order net inflow, and opening large-order net inflow. Large-order inflow and outflow order counts were stronger before 2017, while net-flow measures were steadier; the opening net-flow rate had the strongest reported information ratio and positive-IC frequency. Residual tests suggest order-flow measures retain information beyond traditional price-volume factors, though turnover overlaps substantially. Results cover Chinese equities from 2014 to early 2018, may not persist as market behavior changes, and short holding periods can create high turnover and trading costs.

Key ideas

• Order-flow measures derived from order records capture micro-level supply and demand information in individual stocks.
• The study groups fifty measures into eight categories and tests them with portfolio sorts, regressions, and information coefficients.
• Four factor groups show some stock-selection usefulness, while several active-buying and late-session groups perform weakly.
• Residual analyses indicate incremental information beyond momentum, turnover, and volatility, with notable overlap with turnover.
• The evidence is historical and short holding periods may make transaction costs material.

Tags

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