Portfolio Strategy
CSAM
Walk-forwardCross-Sectional Allocation Model
- Rebalancing
- Monthly
- Universe
- 39 Futures & ETFs
- As of
- ...
The Strategy
CSAM (Cross-Sectional Allocation Model) ranks 39 assets across equity indices, fixed income, energy, metals, agriculture, livestock, currencies, and crypto using a quantitative scoring system. A macro overlay derived from economic conditions adjusts overall portfolio exposure.
Each month, the model predicts risk-adjusted forward returns for 39 assets spanning equity index futures, fixed income, energy, metals, agriculture, livestock, currencies, crypto, and broad ETFs. The top 5 assets with positive predictions are held long, weighted by inverse volatility. A macro overlay scales overall exposure based on the walk-forward FRED composite.
Walk-Forward Test Active Starting June 23, 2026, this strategy is being tested with market data in a walk-forward framework. All historical data before this date represents backtested performance. The vertical marker in the chart indicates the start of the walk-forward period.
How It Works
On the last trading day of each month, the model generates forward-looking scores for all 39 assets. The top 5 assets by score are selected and weighted by inverse realized volatility, concentrating capital in the highest-conviction opportunities.
The dashboard displays both backtested and live performance. The shaded area marks June 23, 2026, the start of the walk-forward test. All data to the left represents historical backtesting; all data to the right represents hypothetical out-of-sample performance. This allows direct comparison of in-sample vs. out-of-sample results.
Simulation Settings
Set your start date and initial capital to configure your portfolio. The CSAM strategy uses monthly Quantitative cross-sectional scoring across 39 assets with macro overlay. Data is available from February 2011.
Model Performance
Drawdown
Peak-to-trough declineHistorical Allocation
Walk-Forward Performance
Since Apr 25, 2026Model Asset Class Allocation
Model ETF Allocation
Model Allocation
Hypothetical model weights for illustration only. Shown as of the latest data refresh. Model value: initial capital plus accumulated hypothetical returns.
| Symbol | Weight | Price | Shares | Value |
|---|---|---|---|---|
| Loading holdings... | ||||
Performance Metrics
Walk-Forward Test Status
Walk-forward testing started April 25, 2026
Monthly rebalancing (next: --)
Crisis Performance
Strategy behavior during major market downturns:
| Crisis | CSAM | SPY |
|---|---|---|
| 2008 GFC | -- | -55% |
| 2020 Covid | -- | -34% |
| 2022 Bear | -- | -25% |
CSAM vs SPY Portfolio
Full backtest period comparison:
| Metric | CSAM | SPY |
|---|---|---|
| CAGR | -- | -- |
| Volatility | -- | -- |
| Sharpe Ratio | -- | -- |
| Max Drawdown | -- | -- |
Statistical Validation
CSAM passed 6 of 7 validation gates in a rigorous multi-phase framework. DSR requires interpretation of independent trial count. Walk-forward FRED selection eliminates look-ahead bias in the macro feature.
200 shuffled prediction sequences tested. None matched the observed Sharpe ratio, confirming the scoring system adds value over random selection.
5,000 block bootstrap samples (21-day blocks) produce a 95% confidence interval for the Sharpe ratio entirely above zero. Based on Politis & Romano (1994).
Walk-forward 10-fold validation. All folds produce positive Sharpe ratios, confirming stability across different time periods.
10% noise injection retains 100% of original Sharpe ratio across 100 seeds. Strategy is robust to small perturbations in the return stream.
Positive returns in 15 of 16 calendar years (2011-2026). Only 2018 showed a small negative return (-10.4%).
Walk-forward 5-fold validation confirms positive Sharpe in all folds. No single period drives the overall result.
Validation based on White (2000), Politis & Romano (1994), Bailey & Lopez de Prado (2014), and Lopez de Prado (2018). All tests conducted on the fixed production strategy. Past statistical performance does not guarantee future results.