Best Component Portfolio (JP Growth Stocks)
Best Component Portfolio (JP Growth Stocks) backtest. Median Sharpe across 3 parameter sets is +1.43. p-value is 0.13. Max DD is -6.5%. Period: 2024-07-01 to 2025-06-30. Best Sharpe +1.63 is not the representative figure.
Verdict: rejected. The median reaches neither promising (Sharpe > 0.5 and p-value < 0.05) nor marginal (Sharpe > 0.05 and best p-value < 0.10). Primary reason: no_edge. It is not regime, frequency, or cost, and the median does not clear the bar.
Performance summary
Hypothesis
A multi-strategy portfolio composed of Phase 2c/2d best variants (ml-factor best version + FLAT-only reversal + fx-carry + pair-trading) achieves higher regime stability than any single component. Phase 2b showed regime-conditional weighting improved UP-regime Sharpe from -1.11 to 0.36.
Why it didn't work
Across 3 grid combinations, median Sharpe=1.43, best=1.63. Primary failure mode: no_edge.
Facts (5)
Runs (3)
| run | params | sharpe | UP | FLAT | DOWN |
|---|---|---|---|---|---|
method=equal_weight r-portfolio-best-components-jp-growth-0001 | method: "equal_weight" | +1.43 | +0.49 | +1.71 | +2.44 |
method=risk_parity r-portfolio-best-components-jp-growth-0004 | method: "risk_parity" | 0.00 | — | — | — |
method=regime_conditional r-portfolio-best-components-jp-growth-0007 | method: "regime_conditional" | +1.63 | +0.57 | +1.52 | +2.85 |