Data Health
Live monitoring of BTC/USDT 4H dataset, ML predictions, and feature coverage.
All Systems Healthy
Total OHLCV Bars
14,122
6.4 years of 4H data
ML Predictions
6,306
35 months coverage
Features Engineered
74
13,719 aligned samples
Data Gaps
0
No gaps > 1 day
Dataset Overview
Source
Binance Spot BTC/USDT 4H
Date Range
2020-01-01 → 2026-06-11
Duration
2,353 days (6.4 years)
Missing Values
0 (all columns complete)
Duplicate Timestamps
0
Monthly Density
65 – 186 bars/month (avg 181)
Yearly Breakdown
| Year | Bars | Start | End | Avg Close | Coverage |
|---|---|---|---|---|---|
| 2020 | 2,195 | Jan 1 | Dec 31 | $11,076 | 100% |
| 2021 | 2,190 | Jan 1 | Dec 31 | $47,355 | 100% |
| 2022 | 2,190 | Jan 1 | Dec 31 | $28,219 | 100% |
| 2023 | 2,190 | Jan 1 | Dec 31 | $28,807 | 100% |
| 2024 | 2,196 | Jan 1 | Dec 31 | $65,904 | 100% |
| 2025 | 2,190 | Jan 1 | Dec 31 | $101,629 | 100% |
| 2026 | 971 | Jan 1 | Jun 11 | $75,521 | 41% |
ML Pipeline Status
LightGBM Predictions
- Coverage: 6,306 bars (2023-02 → 2025-12)
- Missing values: 0
- Regimes: chop 58.3%, crash 16.5%, bear 13.8%, bull 9.9%
Triple-Barrier Labels
- Total labels: 14,117
- Positive (profitable): 5,301 (37.6%)
- Exit: stop-loss 57.5%, take-profit 27.1%, time 15.4%
Feature Matrix
- Shape: 13,719 × 74
- NaN count: 0
- Inf count: 0
- Base features: 97 → 74 after collinearity audit
Alignment
- OHLCV ∩ Predictions: 6,306 bars
- OHLCV ∩ Labels: 14,117 bars
- Predictions ∩ Labels: 6,306 bars
Data Integrity Checks
No missing values
No duplicate timestamps
No gaps > 1 day
No NaN in features
No Inf in features
Full 6.4yr coverage