AI QuantAll research examples

Backtested research example

Sector Relative Strength Rotation (3/6/12m vs SPY)

SECTOR_MOMENTUM
sector-rotationmomentumetfrelative-strengthmonthly-rebalanceus

🗓 Backtest period: 2020-01-01..2026-07-01

StartTotal 329.3%End
Max DD -22.7%

Backtest metrics

Sharpe
1.21
Total Return
159.3%
Max Drawdown
-23.8%
CAGR
15.8%
Volatility
20.4%
Trades
56

Strategy Card

Sector Relative Strength Rotation (3/6/12m vs SPY) — strategy card

Spec ID: spec-sector-momentum-rotation-etf-top50-1767027941_v15 · Generated: 2026-07-10 10:42 UTC

Cluster: Momentum · Sub Cluster: Sector Etf Relative Strength Rotation

One-line description

Monthly rotates into the three most attractive U.S. sector ETFs based on composite 3-, 6-, and 12-month relative strength versus SPY. Positions are equal-weighted, long-only, and traded at the next open after a one-day signal lag.

Why this trade exists

Sector leadership tends to persist over intermediate horizons because macro shocks, earnings revisions, policy regimes, and investor flows affect industries unevenly. Ranking sectors relative to SPY attempts to isolate cross-sectional leadership from the broad equity-market beta component.

Using multiple lookbacks reduces dependence on a single momentum horizon: 3-month strength captures faster rotation, while 6- and 12-month strength emphasize more durable trends. ETF implementation keeps the trade liquid and diversified within each sector, but the edge remains exposed to momentum crashes and regime reversals.

Algorithm

[code omitted from public view]

Parameters

Param Value Notes
Universe Top 50 U.S. ETFs by capitalization Sector ETF filter applied using metadata/name keywords such as Select Sector, Sector, SPDR, XL
Benchmark SPY Relative strength denominator
Lookbacks 63, 126, 252 trading days Approx. 3/6/12 months
Score Mean z-score across lookbacks Cross-sectional normalization by horizon
Selection count 3 ETFs Top-ranked sector ETFs only
Position sizing Equal weight Long-only, gross leverage 1.0
Max weight per position 40% Risk override; effectively non-binding for 3 equal weights unless fewer names
Min positions 3 Strategy expects at least 3 candidates
Rebalance Monthly, first trading day Signals lagged by 1 trading day
Execution Next open Uses prior signal to reduce look-ahead risk
Liquidity filter 63-day ADV >= 250,000 shares Share-volume based liquidity screen
Missing data Drop if missing any lookback Requires 252 trading days of history
Costs 10 bps transaction cost + proportional spread slippage Spread proxy uses 20 days
Stop loss None No explicit downside stop
Volatility targeting False Full exposure maintained when invested
Backtest window 2020-01-01 to 2026-07-01 specified Provided data period is 2018-01-01 to 2023-01-01; verify effective test coverage

Look-ahead audit

# Concern Status
1 Signal uses prices unavailable at rebalance execution Mitigated by 1-trading-day signal lag and next-open execution
2 ETF universe membership may use future metadata or current top-50 list Needs verification; annual point-in-time screening is indicated but survivorship risk remains possible
3 Relative strength calculations may include same-day close before next-open trade Acceptable if signal date precedes execution date after lag
4 Missing data and liquidity filters could accidentally use forward information Should be computed only from trailing history as specified
5 Corporate actions / total return adjustment Requires adjusted total-return series or dividend-adjusted ETF prices
6 Backtest date coverage inconsistency Backtest interval and available data period should be reconciled before relying on live expectations

Caveats / known limitations

Results

The backtest shows strong absolute performance for a simple monthly sector-rotation rule, with total return of 159.28%, Sharpe of 1.21, Sortino of 1.49, and Calmar of 0.66. Risk is material: volatility is 20.39% and max drawdown reaches -23.83%, consistent with concentrated long-only equity-sector exposure. The high win rate of 85.71% and profit factor of 26.94 over 56 trades are encouraging but should be interpreted cautiously given the short effective sample and the need to validate point-in-time universe construction.

Backtest metrics snapshot

Metric Value
Total Return 159.28%
Sharpe 1.21
Sortino 1.49
Calmar 0.66
Max Drawdown -23.83%
Volatility 20.39%
Win Rate 85.71%
Profit Factor 26.94
Total Trades 56
Symbols 50 (AGG, BIL, BIV, BND, BNDX, BSV, DIA, EFA, GLD, IEFA, +40 more)

Backtests are historical simulations for research purposes only. They are not investment advice and do not guarantee future performance.