Files
xau-ai-trading-bot/h1_analysis_output.txt
T
buckybonez f736f1f75b chore: add backtest #39 and HMM investigation artifacts
Added research artifacts from HMM investigation:
- backtest_39_h1_hmm.py — H1 vs M15 HMM comparison attempt
- 39_h1_hmm_results/ — Partial backtest results
- analyze_*.py — ML model and H1 feature analysis scripts
- *_output.txt — Analysis outputs showing HMM degeneracy

Updated:
- data/risk_state.txt — Latest risk state (daily_loss: 18.77, daily_profit: 22.49)

Note: Backtest #39 had import compatibility issues but led to critical
discovery of alternating HMM pattern bug (fixed in c02c2e9).

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-02-09 10:51:49 +07:00

151 lines
8.2 KiB
Plaintext

================================================================================
H1 FEATURE DEEP ANALYSIS
================================================================================
Total features in model: 76
H1 features found: 9
--- ALL H1 FEATURES ---
Rank #68: h1_atr_ratio importance= 0.0000
Rank #66: h1_ema20 importance= 0.0000
Rank # 7: h1_ema20_distance importance= 95.8865
Rank #67: h1_fvg_active importance= 0.0000
Rank #10: h1_market_structure importance= 61.8312
Rank #16: h1_ob_proximity importance= 33.1670
Rank # 6: h1_rsi importance= 112.2508
Rank #17: h1_swing_proximity importance= 30.4503
Rank #12: h1_trend_strength importance= 45.8723
--- TOP 10 H1 FEATURES (by importance) ---
1. Rank # 6: h1_rsi 112.2508
2. Rank # 7: h1_ema20_distance 95.8865
3. Rank # 10: h1_market_structure 61.8312
4. Rank # 12: h1_trend_strength 45.8723
5. Rank # 16: h1_ob_proximity 33.1670
6. Rank # 17: h1_swing_proximity 30.4503
7. Rank # 66: h1_ema20 0.0000
8. Rank # 67: h1_fvg_active 0.0000
9. Rank # 68: h1_atr_ratio 0.0000
--- H1 FEATURE STATISTICS ---
Total H1 features: 9
Mean importance: 42.1620
Median importance: 33.1670
Max importance: 112.2508
Min importance: 0.0000
H1 features in top 10: 3
H1 features in top 20: 6
H1 features in top 30: 6
================================================================================
ALL FEATURE NAMES IN MODEL
================================================================================
1. rsi 36.2503
2. atr 0.0000
3. atr_percent 0.0000
4. macd 65.5672
5. macd_signal 75.9652
6. macd_histogram 0.0000
7. bb_middle 0.0000
8. bb_upper 0.0000
9. bb_lower 0.0000
10. bb_width 0.0000
11. bb_percent_b 21.0445
12. ema_9 0.0000
13. ema_21 21.6908
14. ema_cross_bull 0.0000
15. ema_cross_bear 0.0000
16. volume_sma 0.0000
17. volume_ratio 25.7497
18. volume_increasing 0.0000
19. high_volume 0.0000
20. returns_1 167.9189
21. returns_5 2.9630
22. returns_20 11.5534
23. log_returns 186.0548
24. price_position 39.1339
25. dist_from_sma_20 11.1024
26. volatility_20 0.0000
27. normalized_range 0.0000
28. avg_normalized_range 0.0000
29. close_lag_1 18.1662
30. close_lag_2 18.1221
31. close_lag_3 0.0000
32. close_lag_5 24.4038
33. higher_high 0.0000
34. lower_low 0.0000
35. hh_count_5 0.0000
36. ll_count_5 0.0000
37. hour 14.3622
38. weekday 0.0000
39. london_session 0.0000
40. ny_session 0.0000
41. swing_high 0.0000
42. swing_low 0.0000
43. is_fvg_bull 23.1050
44. is_fvg_bear 0.0000
45. fvg_signal 0.0000
46. ob 417.3543
47. ob_mitigated 127.2658
48. bos 0.0000
49. choch 0.0000
50. market_structure 0.0000
51. regime 0.0000
52. regime_confidence 0.0000
53. h1_ema20 0.0000
54. h1_market_structure 61.8312
55. h1_ema20_distance 95.8865
56. h1_trend_strength 45.8723
57. h1_swing_proximity 30.4503
58. h1_fvg_active 0.0000
59. h1_ob_proximity 33.1670
60. h1_atr_ratio 0.0000
61. h1_rsi 112.2508
62. fvg_gap_size_atr 14.4616
63. fvg_age_bars 0.0000
64. ob_width_atr 40.7235
65. ob_distance_atr 119.3701
66. bos_recency 0.0000
67. confluence_score 0.0000
68. swing_distance_atr 0.0000
69. regime_duration_bars 0.0000
70. regime_transition_prob 0.0000
71. volatility_zscore 19.7360
72. crisis_proximity 15.9215
73. wick_ratio 0.0000
74. body_ratio 0.0000
75. gap_from_prev_close 7.8998
76. consecutive_direction 57.4307
================================================================================
CHECKING TRAINING DATA FOR H1 FEATURES
================================================================================
Dataset columns: 72
H1 columns in dataset: 0
NO H1 COLUMNS IN TRAINING DATA!
Looking for potential H1-related columns:
hour
================================================================================
CHECKING FEATURE ENGINEERING CODE
================================================================================
NO H1 references found in feature_eng.py
================================================================================
CONCLUSION
================================================================================
Traceback (most recent call last):
File "C:\Users\Administrator\Videos\Smart Automatic Trading BOT + AI\analyze_h1_features.py", line 151, in <module>
print(f"\n\u2713 Model HAS {len(h1_features)} H1 features")
~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Python313\Lib\encodings\cp1252.py", line 19, in encode
return codecs.charmap_encode(input,self.errors,encoding_table)[0]
~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
UnicodeEncodeError: 'charmap' codec can't encode character '\u2713' in position 2: character maps to <undefined>