Portfolio Strategy
RCDA
Walk-forwardRegime-Conditional Defensive Allocation — hedge overlay for AMRI
- Rebalancing
- Daily
- Universe
- SPY + Hedge Basket
- As of
- ...
Backtest 2008 to present. Walk-forward test active since May 10, 2026.
The Strategy
RCDA (Regime-Conditional Defensive Allocation) is a defensive overlay for the AMRI model. When AMRI exits to cash, instead of holding idle capital, RCDA identifies the current macroeconomic regime and deploys a tailored basket of defensive assets.
The regime classification distinguishes between four states: expansionary, inflationary, deflationary, and liquidity stress. Each regime has a different optimal composition of hedging instruments. When AMRI is invested in equities, RCDA is inactive and the portfolio simply holds SPY.
Walk-forward test active since May 10, 2026. All historical data before this date represents backtested performance.
How It Works
The overlay monitors daily macro signals. When AMRI moves to cash, RCDA classifies the current regime and allocates across hedging instruments such as long-duration Treasuries, gold, managed futures, TIPS, the dollar index, and short-term bills. Allocation weights shift with the regime.
The dashboard shows the combined performance: AMRI's equity periods plus RCDA's hedge basket during cash periods. The shaded area marks the start of the walk-forward test.
Simulation Settings
Set your start date and initial capital. RCDA overlays the AMRI model with a regime-conditional hedge basket during cash periods.
Model Performance
Drawdown
Peak-to-trough declineHistorical Position
SPY vs Hedge Basket over timeMacro Regime
Regime classification over timeCurrent Signal
Current Allocation
Walk-Forward Test
Crisis Performance
| Crisis | RCDA | SPY |
|---|---|---|
| 2008 GFC | -- | -55% |
| 2020 Covid | -- | -34% |
| 2022 Bear | -- | -25% |
RCDA vs AMRI vs SPY
| Metric | RCDA | AMRI | SPY |
|---|---|---|---|
| CAGR | -- | -- | -- |
| Volatility | -- | -- | -- |
| Sharpe | -- | -- | -- |
| Max DD | -- | -- | -- |
Regime Allocation Matrix
When AMRI signals CASH, RCDA deploys a hedge basket conditioned on the current macro regime:
Detailed Metrics
Statistical Validation
RCDA was validated using a multi-phase robustness framework before deployment. All tests were conducted on the fixed production strategy.
10,000 random position sequences with identical average market exposure. Fewer than 0.3% matched the observed risk-adjusted return.
95% confidence interval for the Sharpe ratio from 10,000 block bootstrap samples (6-month blocks). Entirely above zero.
Probability that observed Sharpe exceeds what would be expected by chance given the number of trials. Based on Bailey & Lopez de Prado (2014).
Combinatorial purged cross-validation across all non-overlapping splits. Median OOS Sharpe of 0.70, all splits positive.
Increasing noise levels produce monotonically decreasing Sharpe ratios, confirming the model captures genuine signal rather than noise artifacts.
Significant improvement in Sharpe ratio and CVaR reduction versus AMRI standalone. The overlay adds measurable risk-adjusted value.
Validation framework based on White (2000), Politis & Romano (1994), and Bailey & Lopez de Prado (2014). All tests were conducted on the fixed production strategy without post-hoc parameter tuning.