Industry Trend Allocation with Cross-Sectional and Time-Series Momentum
Summary
This report studies trend-based allocation across CITIC Securities’ first-level industry indexes. It frames the process as choosing assets, measuring momentum, selecting industries, assigning weights, and managing portfolio tracking. The analysis describes positive long-run returns across the indexes, substantial variation among industries, recurring leadership rotation, and generally high correlations. Autocorrelation tests and comparisons between observation-period and holding-period returns are cited as evidence of momentum over periods of six months or longer.
The portfolio example ranks 29 industries by one-month absolute return, divides them into five groups, and equally weights each group for a one-month holding period. The top-six portfolio had a reported annualized return of 18.17%, outperforming the other groups and the equal-weight market portfolio, but it also had sizable drawdowns and volatility. The report examines alternative momentum measures, selection rules, weights, and rebalancing. Its proposed improvements combine cross-sectional and time-series momentum screening and add stop-losses to limit the effect of sharp reversals within a month. The results are historical simulations, and the summary provides limited detail about implementation costs or robustness beyond the studied sample.
Key ideas
- The report applies a five-step trend-allocation process to industry indexes.
- Industry leadership rotates, while the indexes remain relatively highly correlated.
- Autocorrelation and return comparisons indicate momentum over half-year or longer periods.
- A one-month cross-sectional ranking strategy favored the top six industries but still had high volatility and drawdowns.
- Combining cross-sectional and time-series momentum with stop-losses is presented as a way to improve selection and manage reversals.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.