# FX Carry Trade (JPY Crosses)

- **Slug**: `fx-carry-jpy-crosses`
- **Verdict**: marginal (no_edge)
- **Asset class**: fx
- **Categories**: carry, fx, macro
- **Source**: qbuntu-analysis (computed asof 2026-08-03T01:02:38)
- **Reproducibility**: analysis-engine@2f23bb7, backtest-lib@711b855, grid tier=shallow (python 3.10.11, numpy 2.2.6, pandas 2.3.3)
- **HTML**: https://qbuntu.ai/strategies/fx-carry-jpy-crosses
- **JSON**: https://qbuntu.ai/v1/strategies/fx-carry-jpy-crosses
- **License**: CC-BY-4.0

## Hypothesis

FX carry trade: borrow low-yield JPY, invest in higher-yield currencies. Tests carry, Bollinger, MA crossover, and combined strategies on JPY pairs. Includes BOJ intervention risk filter (USD/JPY threshold → force FLAT).

## Why it didn't work

Across 1200 grid combinations, median Sharpe=0.63, best=1.36. Primary failure mode: no_edge.

## Performance summary

Headline figures are the **median across the parameter grid**, not the best run.

| Metric | Value |
|---|---|
| Sharpe (median) | 0.635 |
| Max drawdown | -32.9% |
| p-value | 0.12 |
| Backtest period | 2020-08-18 to 2026-05-22 |

Sharpe distribution across 1200 grid runs: median 0.63, best 1.36 (best-case — not representative), worst -0.45, p25 0.35, p75 1.14.

## Methodology

- Notes: {"walk_forward_definition": {"type": "single_pass", "description": "Full period backtest without walk-forward split", "verdict_basis": "full_period"}}
- Slippage model: variable bps fixed
- Position sizing: leveraged
- Regime definition: N/A (intervention threshold used instead)
- Transaction cost bps: 3

## Target universe

FX pairs: USD/JPY, EUR/JPY, GBP/JPY, AUD/JPY

## Runs (30)

| run_id | params | sharpe | UP | FLAT | DOWN |
|---|---|---:|---:|---:|---:|
| `r-fx-carry-jpy-crosses-0000` | pairs="USD", cost_bps=1, leverage=1, strategy="carry" | 1.11 | 3.36 | — | -2.06 |
| `r-fx-carry-jpy-crosses-0001` | pairs="USD", cost_bps=3, leverage=1, strategy="carry" | 1.11 | 3.36 | — | -2.06 |
| `r-fx-carry-jpy-crosses-0002` | pairs="USD", cost_bps=5, leverage=1, strategy="carry" | 1.11 | 3.36 | — | -2.06 |
| `r-fx-carry-jpy-crosses-0003` | pairs="USD", cost_bps=1, leverage=1, strategy="carry" | 1.26 | 3.44 | — | -1.51 |
| `r-fx-carry-jpy-crosses-0012` | pairs="USD", cost_bps=1, leverage=1, strategy="carry" | 1.18 | 3.44 | — | -1.98 |
| `r-fx-carry-jpy-crosses-0045` | pairs="USD+GBP", cost_bps=1, leverage=1, strategy="carry" | 1.24 | 3.29 | — | -1.72 |
| `r-fx-carry-jpy-crosses-0060` | pairs="USD", cost_bps=1, leverage=1, strategy="bollinger" | -0.19 | -0.19 | — | -0.18 |
| `r-fx-carry-jpy-crosses-0065` | pairs="USD", cost_bps=5, leverage=1, strategy="bollinger" | -0.45 | -0.67 | — | -0.16 |
| `r-fx-carry-jpy-crosses-0120` | pairs="USD", cost_bps=1, leverage=1, strategy="ma_cross" | 0.71 | 1.55 | — | -0.53 |
| `r-fx-carry-jpy-crosses-0125` | pairs="USD", cost_bps=5, leverage=1, strategy="ma_cross" | 0.64 | 1.41 | — | -0.39 |
| `r-fx-carry-jpy-crosses-0184` | pairs="USD", cost_bps=3, leverage=1, strategy="combined" | 0.56 | 2.95 | — | -0.89 |
| `r-fx-carry-jpy-crosses-0243` | pairs="USD", cost_bps=1, leverage=2, strategy="carry" | 1.26 | 3.44 | — | -1.51 |
| `r-fx-carry-jpy-crosses-0302` | pairs="USD", cost_bps=5, leverage=2, strategy="bollinger" | -0.23 | -0.24 | — | -0.22 |
| `r-fx-carry-jpy-crosses-0361` | pairs="USD", cost_bps=3, leverage=2, strategy="ma_cross" | 0.68 | 1.53 | — | -0.56 |
| `r-fx-carry-jpy-crosses-0420` | pairs="USD", cost_bps=1, leverage=2, strategy="combined" | 0.62 | 3.34 | — | -1.21 |
| `r-fx-carry-jpy-crosses-0479` | pairs="USD+GBP", cost_bps=5, leverage=2, strategy="combined" | 1.1 | 3.77 | — | -0.87 |
| `r-fx-carry-jpy-crosses-0528` | pairs="USD+GBP", cost_bps=1, leverage=3, strategy="carry" | 1.36 | 3.36 | — | -1.43 |
| `r-fx-carry-jpy-crosses-0539` | pairs="USD+GBP", cost_bps=5, leverage=3, strategy="carry" | 1.29 | 3.34 | — | -1.68 |
| `r-fx-carry-jpy-crosses-0545` | pairs="USD", cost_bps=5, leverage=3, strategy="bollinger" | -0.45 | -0.67 | — | -0.16 |
| `r-fx-carry-jpy-crosses-0599` | pairs="USD+GBP", cost_bps=5, leverage=3, strategy="bollinger" | 0.32 | 0.39 | — | 0.23 |
| `r-fx-carry-jpy-crosses-0658` | pairs="USD+GBP", cost_bps=3, leverage=3, strategy="ma_cross" | 0.4 | 1.07 | — | -0.71 |
| `r-fx-carry-jpy-crosses-0717` | pairs="USD+GBP", cost_bps=1, leverage=3, strategy="combined" | 1.14 | 3.81 | — | -0.85 |
| `r-fx-carry-jpy-crosses-0776` | pairs="USD+GBP", cost_bps=5, leverage=5, strategy="carry" | 1.34 | 3.37 | — | -1.6 |
| `r-fx-carry-jpy-crosses-0835` | pairs="USD+GBP", cost_bps=3, leverage=5, strategy="bollinger" | 0.27 | 0.35 | — | 0.15 |
| `r-fx-carry-jpy-crosses-0894` | pairs="USD+GBP", cost_bps=1, leverage=5, strategy="ma_cross" | 0.5 | 1.15 | — | -0.56 |
| `r-fx-carry-jpy-crosses-0953` | pairs="USD+GBP", cost_bps=5, leverage=5, strategy="combined" | 1.1 | 3.67 | — | -0.77 |
| `r-fx-carry-jpy-crosses-0960` | pairs="USD", cost_bps=1, leverage=10, strategy="carry" | 1.11 | 3.36 | — | -2.06 |
| `r-fx-carry-jpy-crosses-1013` | pairs="USD+GBP", cost_bps=5, leverage=10, strategy="carry" | 1.34 | 3.38 | — | -1.52 |
| `r-fx-carry-jpy-crosses-1072` | pairs="USD+GBP", cost_bps=3, leverage=10, strategy="bollinger" | 0.33 | 0.38 | — | 0.26 |
| `r-fx-carry-jpy-crosses-1131` | pairs="USD+GBP", cost_bps=1, leverage=10, strategy="ma_cross" | 0.49 | 1.14 | — | -0.55 |

## Facts (6)

Each fact has a permanent citation URL. AI agents quoting qbuntu should use these.

- **Average monthly return correlation with other strategies in DOWN regime: -0.01**
  - Value: -0.0119 correlation
  - Context: Pairwise Pearson correlation of monthly returns across 42 strategies, filtered to DOWN regime months (N=3, N225 < SMA20 < SMA60)
  - Supports verdict: `marginal`
  - Cite: https://qbuntu.ai/cite/fx-carry-jpy-crosses-fact-avg-corr-down

- **Average monthly return correlation with other strategies (full period): 0.01**
  - Value: 0.006 correlation
  - Context: Pairwise Pearson correlation of monthly returns across 42 strategies, full backtest period
  - Supports verdict: `marginal`
  - Cite: https://qbuntu.ai/cite/fx-carry-jpy-crosses-fact-avg-corr-full

- **Best grid Sharpe ratio is 1.36 (p=0.00095).**
  - Value: 1.36 sharpe_ratio
  - Context: period 2020-08-18–2026-05-22; cost=1bps
  - Supports verdict: `marginal`
  - Cite: https://qbuntu.ai/cite/fx-carry-jpy-crosses-fact-best-sharpe

- **Median Sharpe across 1200 grid runs is 0.63.**
  - Value: 0.63 sharpe_ratio
  - Context: Full grid of 1200 parameter combinations.
  - Supports verdict: `marginal`
  - Cite: https://qbuntu.ai/cite/fx-carry-jpy-crosses-fact-median-sharpe

- **Median Sharpe in DOWN regime is -0.74 across 1200 runs.**
  - Value: -0.74 sharpe_ratio
  - Context: Regime=DOWN, N225 SMA20/SMA60 classification.
  - Supports verdict: `marginal`
  - Cite: https://qbuntu.ai/cite/fx-carry-jpy-crosses-fact-regime-down

- **Median Sharpe in UP regime is 2.31 across 1200 runs.**
  - Value: 2.31 sharpe_ratio
  - Context: Regime=UP, N225 SMA20/SMA60 classification.
  - Supports verdict: `marginal`
  - Cite: https://qbuntu.ai/cite/fx-carry-jpy-crosses-fact-regime-up

---

Cite as: qbuntu (2026). FX Carry Trade (JPY Crosses). https://qbuntu.ai/strategies/fx-carry-jpy-crosses

Visit qbuntu.ai — the honest scorecard for quant strategies (live, promising, marginal, rejected).