Feature Iteration Run #986: zz_sylvain

# AUTOLOOP: zz_sylvain (8 iterations)

> **Phase**: Y4.7 Option B (DSL → Python code → FeatureSet candidate)
> **Target indicator**: zz_sylvain
> **Base feature_set_id**: 7
> **Claim**: (default) Propose a new k_* feature that improves risk-adjusted returns
> **Objective mode**: scalarize
> **Metrics**: ['profit_factor', 'sharpe_ratio', 'win_rate', 'max_drawdown', 'test_accuracy']
> **Total duration**: 222.41s
> **Total accepted**: 0
> **Total rejected**: 8

## Iterations 詳情

### Iteration 1
- DSL: `obv_ensemble_3 * (1 - hl_range) / (1 + k_momentum_pct_change.abs() * 10)` (feature=k_loop_iter_1_2ad30e)
- DSL valid: True
- FeatureSet candidate id: 1260
- baseline PF: 1.947 → modified PF: 1.085
- t_stat: 0.000, p_value: 0.463, cohens_d: 0.243
- scalar_score: -0.4789
- accepted: False, reason: normalized_score -0.2581 <= 0.0, sharpe_delta -16.8215 < -0.1, hypothesis test not confirmed

**Metrics summary:**
  - profit_factor: baseline=1.947, modified=1.085, diff=-0.862
  - sharpe_ratio: baseline=19.162, modified=2.341, diff=-16.821
  - win_rate: baseline=0.600, modified=0.500, diff=-0.100
  - max_drawdown: baseline=0.056, modified=0.187, diff=0.131
  - test_accuracy: baseline=0.992, modified=0.992, diff=-0.000

### Iteration 2
- DSL: `k_high.pct_change(12) * (1 + hl_range) * (1 - price_position) * (1 - dist_to_high)` (feature=k_iter2_k_high_pct_change_hl_range_price_position)
- DSL valid: False
- FeatureSet candidate id: None
- baseline PF: 0.000 → modified PF: 0.000
- t_stat: 0.000, p_value: 1.000, cohens_d: 0.000
- scalar_score: 0.0000
- accepted: False, reason: quality_gate: C8 leakage: ['dist_to_high']; canonical duplicate: k_high.pct_change * (1 + hl_range) * (1 - price_position) * (1 - dist_to_high)

### Iteration 3
- DSL: `i_atr(k_high, k_low, k_close, 14) / k_close * (1 + abs(o_div_binary(48))) * (1 + k_momentum_pct_change.abs())` (feature=k_iter3_i_atr_k_high_k_low_k_close)
- DSL valid: False
- FeatureSet candidate id: None
- baseline PF: 0.000 → modified PF: 0.000
- t_stat: 0.000, p_value: 1.000, cohens_d: 0.000
- scalar_score: 0.0000
- accepted: False, reason: safe_eval DSL: 變數未定義: abs

### Iteration 4
- DSL: `k_high.pct_change(12) * (1 + hl_range) * (1 - dist_to_low) / (1 + atr_norm)` (feature=k_iter4_k_high_pct_change_hl_range_dist_to_low)
- DSL valid: False
- FeatureSet candidate id: None
- baseline PF: 0.000 → modified PF: 0.000
- t_stat: 0.000, p_value: 1.000, cohens_d: 0.000
- scalar_score: 0.0000
- accepted: False, reason: quality_gate: C8 leakage: ['dist_to_low']

### Iteration 5
- DSL: `i_macd(k_close, 12, 26, 9) * o_pct_change_12 / (1 + atr_norm * 10)` (feature=k_loop_iter_5_5bfa21)
- DSL valid: True
- FeatureSet candidate id: 1261
- baseline PF: 1.947 → modified PF: 1.905
- t_stat: 0.000, p_value: 0.548, cohens_d: 0.213
- scalar_score: -0.0224
- accepted: False, reason: normalized_score -0.0313 <= 0.0, sharpe_delta -0.7047 < -0.1, hypothesis test not confirmed

**Metrics summary:**
  - profit_factor: baseline=1.947, modified=1.905, diff=-0.043
  - sharpe_ratio: baseline=19.162, modified=18.458, diff=-0.705
  - win_rate: baseline=0.600, modified=0.462, diff=-0.138
  - max_drawdown: baseline=0.056, modified=0.101, diff=0.045
  - test_accuracy: baseline=0.992, modified=0.988, diff=-0.004

### Iteration 6
- DSL: `k_high.pct_change(12) * i_rsi(k_close, 14) / (1 + atr_norm * 10) * (1 - price_position)` (feature=k_iter6_k_high_pct_change_i_rsi_k_close)
- DSL valid: False
- FeatureSet candidate id: None
- baseline PF: 0.000 → modified PF: 0.000
- t_stat: 0.000, p_value: 1.000, cohens_d: 0.000
- scalar_score: 0.0000
- accepted: False, reason: quality_gate: NaN ratio 1.000 exceeds 0.3

### Iteration 7
- DSL: `k_momentum_pct_change * o_pct_change_12 / (1 + atr_norm * 10) * (1 - price_position)` (feature=k_loop_iter_7_ef5100)
- DSL valid: True
- FeatureSet candidate id: 1262
- baseline PF: 1.947 → modified PF: 1.085
- t_stat: 0.000, p_value: 0.463, cohens_d: 0.243
- scalar_score: -0.4789
- accepted: False, reason: normalized_score -0.2581 <= 0.0, sharpe_delta -16.8215 < -0.1, hypothesis test not confirmed

**Metrics summary:**
  - profit_factor: baseline=1.947, modified=1.085, diff=-0.862
  - sharpe_ratio: baseline=19.162, modified=2.341, diff=-16.821
  - win_rate: baseline=0.600, modified=0.500, diff=-0.100
  - max_drawdown: baseline=0.056, modified=0.187, diff=0.131
  - test_accuracy: baseline=0.992, modified=0.992, diff=-0.000

### Iteration 8
- DSL: `(k_momentum_pct_change - o_div_amplitude_12) / (1 + atr_norm * 10) * (1 - price_position)` (feature=k_loop_iter_8_c2380e)
- DSL valid: True
- FeatureSet candidate id: 1263
- baseline PF: 1.947 → modified PF: 0.775
- t_stat: 0.000, p_value: 0.278, cohens_d: 0.577
- scalar_score: -3.6248
- accepted: False, reason: normalized_score -0.4195 <= 0.0, sharpe_delta -26.7167 < -0.1, hypothesis test not confirmed, constraints: ['sharpe_ratio >= 1.0: got -7.5544']

**Metrics summary:**
  - profit_factor: baseline=1.947, modified=0.775, diff=-1.172
  - sharpe_ratio: baseline=19.162, modified=-7.554, diff=-26.717
  - win_rate: baseline=0.600, modified=0.424, diff=-0.176
  - max_drawdown: baseline=0.056, modified=0.131, diff=0.075
  - test_accuracy: baseline=0.992, modified=0.992, diff=-0.001