Skip to main content

Checking access...

Portfolio Strategy

AMRI

Walk-forward

Adaptive Macro Regime Index

Rebalancing
Daily
Universe
SPY / Cash
As of
...

The Strategy

AMRI (Adaptive Macro Regime Index) is a systematic strategy that manages exposure to US large-cap equities. It holds a long position in the S&P 500 when its regime assessment is constructive and moves fully to cash when that assessment turns defensive.

The result is a long or flat profile that aims to take part in sustained uptrends while stepping aside during the most severe drawdowns. The model is fully rules-based with no discretionary input. It accepts giving up some upside during extended rallies in return for reducing exposure to large losses.

Walk-Forward Test Active Starting December 14, 2025, this strategy is being tested with market data in a walk-forward framework. All historical data before this date represents backtested performance. The shaded region in the chart indicates the start of the walk-forward period.

How It Works

AMRI runs on a quantitative model that assesses the broad market and macro environment and translates it into a single long or flat decision.

The model is evaluated once per trading day and is built to avoid look-ahead bias, so historical and live results stay directly comparable. The dashboard shows both backtested and out-of-sample performance, with the shaded area marking December 14, 2025, the start of the walk-forward test.

SPY / Cash
Universe
Binary
Long / Flat Signal
Daily
Rebalancing

Simulation Settings

Set your start date and initial capital to configure the model simulation. AMRI evaluates its signal daily. The backtest begins January 2005. Minimum capital: $20,000.

Model Performance

Drawdown

Peak-to-trough decline

Position History

SPY vs Cash over time
Model Value
--
--
YTD Return
--
SPY: --
Volatility
--
Sharpe Ratio
--
Excess return basis

Walk-Forward Performance

Since Dec 14, 2025
--
AMRI Return
--
SPY B&H Return
--
vs SPY B&H
--
Days Live

Current Signal

Signal --
Position --
Days in Position --

Walk-Forward Status

Walk-forward testing started December 14, 2025

Daily rebalancing (signal evaluated each trading day)

Performance Metrics

Sortino Ratio
--
Downside risk-adjusted
Max Drawdown
--
Peak to trough
Calmar Ratio
--
Return/Max DD
Beta
--
vs SPY
VaR (95%)
--
Daily at risk
Exposure
--
Time in market
Up Capture
--
Bull market
Down Capture
--
Bear market
Win Rate
--
Profitable trades
Profit Factor
--
Gross profit / loss

Crisis Performance

Strategy behavior during major market downturns:

Crisis AMRI SPY
2008 GFC---55%
2020 Covid---34%
2022 Bear---25%

AMRI vs SPY Buy & Hold

Full backtest period comparison:

Metric AMRI SPY
Total Return----
CAGR----
Volatility----
Sharpe Ratio----
Max Drawdown----
Exposure--100%

Strategy Health Monitor

Live drift detection and alpha decay tracking

Drift Detection

Detects structural shifts in performance

...
Drift level 0%
0% 50% watch 80% warn 100% alert

Alpha Decay Monitor

Risk-adjusted return, 12 vs 36 months

...
12M Sharpe
--
36M Sharpe
--
12M Alpha
--
Decay Slope
--

The left panel flags structural shifts in performance against expectations. The right panel compares recent risk-adjusted returns with the longer-run track record. A weakening trend points to a fading edge.

Statistical Validation

Before deployment, AMRI was validated with a battery of standard statistical robustness tests. All tests used the fixed production strategy without post-hoc parameter tuning.

Permutation Test passed
p < 0.001

10,000 random position sequences. None matched the observed risk-adjusted return.

Bootstrap passed
CI: 0.95 – 1.71

95% Sharpe ratio confidence interval from 10,000 bootstrap samples. Entirely above zero.

Parameter Robustness passed
100%

180 parameter combinations tested. All produced positive Sharpe ratios. Deployed at the 68th percentile.

Regime Stability passed
All regimes

Positive risk-adjusted returns across bull/bear markets and high/low volatility environments.

Tail Risk passed
−4.5%

CVaR (95th percentile) per month vs −9.7% for SPY. Worst observed month: −6.7%.

Data Integrity verified
Verified

Macro inputs use a publication delay and signals act on the prior close, so the backtest reflects information available at each decision.

All tests conducted on the fixed production strategy. Past statistical performance does not guarantee future results.