Improving A-Share Momentum Factors with Trend, Neutralization, and Bar Design
Summary
This research summary examines ways to improve equity momentum signals in China’s A-share market, where the authors report that reversal has historically been stronger than momentum. It compares conventional momentum with four adaptations: using moving averages to capture the path of prices, neutralizing liquidity-related effects, using residual returns to reduce market-style exposures, and constructing momentum from bars grouped by traded value rather than time. The stated evaluation period runs from 2006 through May 2019.
The reported findings are mixed. Moving-average trend factors show stronger monotonicity, though their information coefficients are weaker than those of conventional momentum. Removing liquidity effects improves monotonicity and raises the reported annualized long-short return. Residual momentum has a slightly lower average information coefficient but less variation, while traded-value-bar momentum has relatively strong long-side returns. These are historical model results from the summary, not guarantees of future performance; the underlying report is not reproduced here, and the authors explicitly warn that models may fail.
Key ideas
- The report describes stronger historical reversal than momentum in the A-share market.
- Moving-average-based trend momentum improves monotonicity but has weaker predictive statistics than conventional momentum.
- Cross-sectional neutralization of liquidity effects improves the reported momentum factor results.
- Residual momentum reduces information-coefficient variability while slightly lowering its average.
- Momentum built from traded-value bars shows a return contribution concentrated more heavily in the long portfolio.
- The findings cover a historical sample and may not persist.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.