Dynamic Hedge Ratio Pairs
Uses Kalman Filter to dynamically estimate the hedge ratio between two assets. More adaptive than rolling OLS regression. Trades spread z-score with time-varying parameters.
At a glance
What the backtest shows
- Beat buy-and-hold on 0 of 1 markets by annual return over 5 years.
- Had a smaller maximum drawdown than buy-and-hold on 1 of 1 markets.
- Best market by Sharpe ratio: SPY (Sharpe -0.21, CAGR -2.0%).
Equity curve
Score and risk profile
Each bar is a percentile against every strategy on the same market and window (100 = best).
How it held up in sell-offs (SPY)
| Period | Strategy | Buy & hold |
|---|---|---|
| Q4 2018 sell-off | -8.6% | -13.9% |
| COVID crash (Feb–Mar 2020) | -7.4% | -25.5% |
| 2022 bear market | -7.2% | -9.1% |
| Market above its 200-day average (annualized) · 67% | +0.2% | +20.1% |
| Market below its 200-day average (annualized) · 33% | +2.1% | +8.7% |
Sell-off dates follow the US stock market. Above/below the 200-day average uses the previous day's position, so there is no look-ahead.
Out-of-sample check (SPY)
| Check | Value |
|---|---|
| Chosen parameters on their 3-year training windows (average Sharpe) | 0.25 |
| Same parameters in the following 6 months, when traded (Sharpe) | 0.03 |
| Default parameters over the same months (Sharpe) | -0.10 |
| Parameter re-selections / changes | 12 / 3 |
| Parameter settings with a positive Sharpe (full history) | 6 of 9 |
| Sharpe across settings: worst / median / best | -0.04 / 0.05 / 0.33 |
Out-of-sample periods start on 2020-11-09. Training-window figures are flattered by picking the best of 9 settings; the out-of-sample figure is what the strategy actually earned. A large drop, or a result that depends on one setting, points to overfitting. How walk-forward testing works.
Results by market (5 years)
| Market | CAGR | Buy & hold CAGR | Sharpe | Max drawdown | Buy & hold max DD | Volatility | Arena Score |
|---|---|---|---|---|---|---|---|
| SPYSPDR S&P 500 | -2.0% | 9.1% | -0.21 | -19.6% | -25.1% | 8.0% | 10 |
Strategy metrics use daily returns without trading costs. Buy-and-hold figures are for the same market and window. Rows are sorted by Sharpe ratio.
SPY across time windows
Rules
Entry
Z < -1.5: long spread (long Y, short X) Z > 1.5: short spread
Exit
Z crosses 0 or reverses
Indicators
Kalman-filtered hedge ratio Dynamic spread Spread z-score
Inputs
Q = 0.001 (process noise) R = 1.0 (measurement noise) Entry_Z = 1.5 Exit_Z = 0.3
Position sizing
Equal dollar weights adjusted by hedge ratio
Risk controls
Max holding period Hedge ratio stability check
Strategy evidence
- SPY · Inspect evidence 2026-11-10
Only available, validated bundles appear below. Missing evidence is not a passing validation result.
Evidence coverage
1 / 1 declared strategy–market pairs validated
Notes
- Kalman Filter adapts to structural changes
- No need for fixed lookback window
- More responsive than OLS-based pairs
Related strategies
Embed this score
Add the live badge to a blog post, README or forum signature. It updates with the daily data.
<a href="https://quantarenas.com/strategy/stat-10-dynamic-hedge-ratio-pairs/"><img src="https://quantarenas.com/badge/stat-10-dynamic-hedge-ratio-pairs.svg" alt="Dynamic Hedge Ratio Pairs Arena Score on Quant Arena" height="20"></a>
[](https://quantarenas.com/strategy/stat-10-dynamic-hedge-ratio-pairs/)
Backtests are historical simulations on daily data and do not include commissions or slippage. They do not guarantee future results and are not investment advice. See the methodology for how every strategy is tested.