Home / Ideas / Cross-asset relative-strength rotation with a correlation short hedge
Cross-asset relative-strength rotation with a correlation short hedge
own the strongest sleeves, short the weakest
A practitioner-validated QuantConnect study ranks ~13 equity/bond/commodity/currency ETFs monthly by average 3/6/9/12-month return, holds the top four equal-weighted, and shorts the single worst at 30% weight — a relative-strength cross-asset rotation distinct from the absolute-trend idea in #1. Reported Sharpe ~0.498 (2007-2026), and the short hedge dampens beta in risk-off regimes.
Why it's relevant now
With 2026 institutional views split (Morgan Stanley favoring equities over bonds, JPMorgan seeing sticky inflation and no cuts), cross-asset dispersion is elevated — the exact condition where relative-strength ranking beats picking a single macro thesis.
Universe
SPY, EFA, EEM, TLT, IEF, LQD, HYG, GLD, DBC, UUP and similar liquid cross-asset ETFs (~10-13 names).
How it works
Each month, rank the sleeve by blended 3/6/9/12-month return, hold the top 4 equal-weighted long and short the bottom-ranked ETF at ~30% weight, rebalance monthly.
Expected performance
Research-derived Sharpe estimate: 0.35–0.6. reported 0.498; realistic after costs given monthly turnover across many ETFs).
Backtest this idea with SignalChain
This is a research lead — not a finished backtest. SignalChain takes an idea like this and runs the whole pipeline inside Claude Code: it researches the concept against academic and practitioner sources, sets benchmarks, writes and lints a VectorBT backtest, runs it, and grades the result PASS/FAIL. One command:
/signalchain Monthly cross-asset ETF rotation: rank SPY EFA EEM TLT IEF LQD HYG GLD DBC UUP by blended 3/6/9/12-month return, hold top 4 long and short the worst
Get SignalChain — $49 →
Research & sources
- quantconnect.com — quantconnect.com
- quantpedia.com — quantpedia.com
- morganstanley.com — morganstanley.com
More macro, rates, and cross-asset strategies ideas
Not financial advice. This page describes a research idea, not a recommendation. Any performance figures are hypothetical, research-derived estimates and are not indicative of future results. SignalChain is a research and educational tool; you are solely responsible for any decisions you make.