# Volume Anomaly (JP Growth Stocks)

- **Slug**: `volume-anomaly-jp-growth`
- **Verdict**: rejected (regime_dependent)
- **Asset class**: equity
- **Categories**: anomaly, factor, volume
- **Source**: qbuntu-analysis (computed asof 2026-07-28T01:10:32)
- **Reproducibility**: analysis-engine@bfafe7d, backtest-lib@711b855, grid tier=shallow (python 3.10.11, numpy 2.2.6, pandas 2.3.3)
- **HTML**: https://qbuntu.ai/strategies/volume-anomaly-jp-growth
- **JSON**: https://qbuntu.ai/v1/strategies/volume-anomaly-jp-growth
- **License**: CC-BY-4.0

## Hypothesis

Campbell, Grossman, Wang (1993): abnormal volume spikes predict subsequent price drift or reversal. High volume + positive return signals momentum continuation; high volume + negative return signals overreaction. Tests momentum, reversal, and volume-only directions.

## Why it didn't work

Across 689 grid combinations, median Sharpe=-0.03, best=0.84. Primary failure mode: regime_dependent.

## Performance summary

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

| Metric | Value |
|---|---|
| Sharpe (median) | -0.0322 |
| Max drawdown | -44.1% |
| p-value | 0.58 |
| Backtest period | 2022-02-01 to 2026-06-01 |

Sharpe distribution across 689 grid runs: median -0.03, best 0.84 (best-case — not representative), worst -1.48, p25 -0.29, p75 0.23.

## Methodology

- Notes: {"walk_forward_definition": {"type": "single_pass", "description": "Full period backtest without walk-forward split", "verdict_basis": "full_period"}}
- Slippage model: open-on-open + variable bps fixed
- Position sizing: equal_weight
- Regime definition: N225 SMA20/SMA60 three-class
- Transaction cost bps: 5

## Target universe

Tokyo Stock Exchange Growth Market, ~495 stocks, tv >= 30M JPY

## Runs (30)

| run_id | params | sharpe | UP | FLAT | DOWN |
|---|---|---:|---:|---:|---:|
| `r-volume-anomaly-jp-growth-0000` | cost_bps=5, direction="momentum", regime_gate=false, holding_days=1 | -0.21 | -0.68 | -0.02 | -1.19 |
| `r-volume-anomaly-jp-growth-0001` | cost_bps=5, direction="momentum", regime_gate=true, holding_days=1 | -0.44 | -1.06 | -0.02 | -1.19 |
| `r-volume-anomaly-jp-growth-0002` | cost_bps=10, direction="momentum", regime_gate=false, holding_days=1 | -0.42 | -0.89 | -0.22 | -1.64 |
| `r-volume-anomaly-jp-growth-0003` | cost_bps=10, direction="momentum", regime_gate=true, holding_days=1 | -0.59 | -1.16 | -0.22 | -1.64 |
| `r-volume-anomaly-jp-growth-0008` | cost_bps=5, direction="vol_only", regime_gate=false, holding_days=1 | -1.19 | -1.45 | -1.1 | -2.13 |
| `r-volume-anomaly-jp-growth-0036` | cost_bps=5, direction="momentum", regime_gate=false, holding_days=20 | 0.41 | 0.1 | 0.61 | -0.77 |
| `r-volume-anomaly-jp-growth-0037` | cost_bps=5, direction="momentum", regime_gate=true, holding_days=20 | 0.31 | 0.01 | 0.61 | -0.77 |
| `r-volume-anomaly-jp-growth-0069` | cost_bps=5, direction="vol_only", regime_gate=true, holding_days=5 | -0.19 | 0.86 | -0.34 | -2.51 |
| `r-volume-anomaly-jp-growth-0101` | cost_bps=5, direction="reversal", regime_gate=true, holding_days=1 | -0.65 | -0.98 | -0.46 | -2.79 |
| `r-volume-anomaly-jp-growth-0133` | cost_bps=5, direction="momentum", regime_gate=true, holding_days=20 | 0.16 | 0.79 | -0.27 | -0.86 |
| `r-volume-anomaly-jp-growth-0144` | cost_bps=5, direction="momentum", regime_gate=false, holding_days=1 | -0.08 | -0.28 | -0.03 | 0.25 |
| `r-volume-anomaly-jp-growth-0171` | cost_bps=10, direction="momentum", regime_gate=true, holding_days=10 | 0.06 | 0.8 | -0.6 | -1.16 |
| `r-volume-anomaly-jp-growth-0184` | cost_bps=5, direction="reversal", regime_gate=false, holding_days=20 | 0.84 | 0.02 | 0.94 | 2.48 |
| `r-volume-anomaly-jp-growth-0206` | cost_bps=10, direction="momentum", regime_gate=false, holding_days=5 | -0.11 | 0.32 | -0.26 | -2.55 |
| `r-volume-anomaly-jp-growth-0238` | cost_bps=10, direction="vol_only", regime_gate=false, holding_days=20 | -0.34 | 0.17 | -0.41 | -2.19 |
| `r-volume-anomaly-jp-growth-0250` | cost_bps=10, direction="vol_only", regime_gate=false, holding_days=1 | -1.48 | -1.23 | -1.43 | -3.83 |
| `r-volume-anomaly-jp-growth-0271` | cost_bps=10, direction="reversal", regime_gate=true, holding_days=10 | 0.13 | 0.06 | 0.25 | -1.11 |
| `r-volume-anomaly-jp-growth-0307` | cost_bps=10, direction="reversal", regime_gate=true, holding_days=5 | 0.01 | 0.52 | -0.18 | -1.4 |
| `r-volume-anomaly-jp-growth-0343` | cost_bps=10, direction="reversal", regime_gate=true, holding_days=1 | 0.19 | 0.4 | 0.11 | -0.31 |
| `r-volume-anomaly-jp-growth-0391` | cost_bps=10, direction="reversal", regime_gate=true, holding_days=1 | 0.14 | -0.24 | 0.38 | 0.09 |
| `r-volume-anomaly-jp-growth-0423` | cost_bps=10, direction="momentum", regime_gate=true, holding_days=20 | 0.18 | 0.22 | 0.26 | -3.46 |
| `r-volume-anomaly-jp-growth-0455` | cost_bps=10, direction="vol_only", regime_gate=true, holding_days=5 | -0.75 | 0.6 | -0.98 | -3.96 |
| `r-volume-anomaly-jp-growth-0487` | cost_bps=10, direction="reversal", regime_gate=true, holding_days=1 | -0.19 | -0.84 | 0.1 | 0.24 |
| `r-volume-anomaly-jp-growth-0527` | cost_bps=10, direction="vol_only", regime_gate=true, holding_days=20 | -0.29 | -0.17 | -0.11 | -5.33 |
| `r-volume-anomaly-jp-growth-0571` | cost_bps=10, direction="reversal", regime_gate=true, holding_days=20 | 0.59 | 1.12 | 0.39 | 0.46 |
| `r-volume-anomaly-jp-growth-0576` | cost_bps=5, direction="momentum", regime_gate=false, holding_days=1 | -0.66 | 0.05 | -0.95 | 0.13 |
| `r-volume-anomaly-jp-growth-0608` | cost_bps=5, direction="vol_only", regime_gate=false, holding_days=10 | -0.21 | 0.44 | -0.35 | -3.16 |
| `r-volume-anomaly-jp-growth-0644` | cost_bps=5, direction="vol_only", regime_gate=false, holding_days=5 | -0.81 | 0.58 | -1.17 | -3.01 |
| `r-volume-anomaly-jp-growth-0676` | cost_bps=5, direction="reversal", regime_gate=false, holding_days=1 | 0.12 | -0.83 | 0.56 | -0.8 |
| `r-volume-anomaly-jp-growth-0720` | cost_bps=5, direction="momentum", regime_gate=false, holding_days=1 | -0.81 | -0.16 | -1.04 | -0.26 |

## Facts (7)

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

- **Average monthly return correlation with other strategies in DOWN regime: 0.07**
  - Value: 0.0743 correlation
  - Context: Pairwise Pearson correlation of monthly returns across 42 strategies, filtered to DOWN regime months (N=3, N225 < SMA20 < SMA60)
  - Supports verdict: `rejected`
  - Cite: https://qbuntu.ai/cite/volume-anomaly-jp-growth-fact-avg-corr-down

- **Average monthly return correlation with other strategies (full period): 0.13**
  - Value: 0.1311 correlation
  - Context: Pairwise Pearson correlation of monthly returns across 42 strategies, full backtest period
  - Supports verdict: `rejected`
  - Cite: https://qbuntu.ai/cite/volume-anomaly-jp-growth-fact-avg-corr-full

- **Best grid Sharpe ratio is 0.84 (p=0.096).**
  - Value: 0.84 sharpe_ratio
  - Context: period 2022-02-01–2026-06-01; cost=5bps
  - Supports verdict: `rejected`
  - Cite: https://qbuntu.ai/cite/volume-anomaly-jp-growth-fact-best-sharpe

- **Median Sharpe across 689 grid runs is -0.03.**
  - Value: -0.03 sharpe_ratio
  - Context: Full grid of 689 parameter combinations.
  - Supports verdict: `rejected`
  - Cite: https://qbuntu.ai/cite/volume-anomaly-jp-growth-fact-median-sharpe

- **Median Sharpe in DOWN regime is -0.86 across 689 runs.**
  - Value: -0.86 sharpe_ratio
  - Context: Regime=DOWN, N225 SMA20/SMA60 classification.
  - Supports verdict: `rejected`
  - Cite: https://qbuntu.ai/cite/volume-anomaly-jp-growth-fact-regime-down

- **Median Sharpe in FLAT regime is -0.16 across 689 runs.**
  - Value: -0.16 sharpe_ratio
  - Context: Regime=FLAT, N225 SMA20/SMA60 classification.
  - Supports verdict: `rejected`
  - Cite: https://qbuntu.ai/cite/volume-anomaly-jp-growth-fact-regime-flat

- **Median Sharpe in UP regime is 0.27 across 689 runs.**
  - Value: 0.27 sharpe_ratio
  - Context: Regime=UP, N225 SMA20/SMA60 classification.
  - Supports verdict: `rejected`
  - Cite: https://qbuntu.ai/cite/volume-anomaly-jp-growth-fact-regime-up

---

Cite as: qbuntu (2026). Volume Anomaly (JP Growth Stocks). https://qbuntu.ai/strategies/volume-anomaly-jp-growth

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