Separating A-Share Momentum and Reversal Signals by Return Components

Summary

This research summary examines why Chinese A-share returns show short-term reversal even though medium- and long-term raw returns do not show the momentum commonly reported in overseas markets. It decomposes stock returns into overnight returns and intraday returns, then splits intraday returns into moderate and extreme components using five-minute data and an outlier-treatment approach. Using Jegadeesh and Titman’s portfolio construction method, it reports momentum in overnight and moderate intraday returns, while extreme intraday returns show pronounced reversal.

The summary says positive extreme intraday moves occur slightly more often and are slightly larger than negative ones, and that extreme moves contribute substantially to stock gains. The one-month reversal component and nine-month momentum component show reported cross-sectional predictive statistics, with different strength across market-cap groups; the one-month signal decays quickly, while the nine-month momentum component has a longer half-life. These findings are research conclusions summarized from a report, not a complete replication package. The text offers limited detail on sample construction, costs, and implementation, so reported signal strength should not be read as net strategy performance.

Key ideas

• The analysis separates stock returns into overnight, moderate intraday, and extreme intraday components.
• Overnight and moderate intraday returns exhibit momentum, while extreme intraday returns exhibit reversal.
• The short-horizon reversal signal decays quickly, while the longer-horizon momentum signal persists longer.
• The reported findings describe predictive factor behavior and do not establish net returns after implementation costs.

Tags

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