File size: 5,610 Bytes
8346b41 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 | """
Convert 1-minute data to higher timeframes (5m, 15m, 1h, 4h, daily).
This allows testing the same strategy on different timeframes to find
the optimal trading interval.
"""
import pandas as pd
import os
import sys
sys.path.append(os.path.dirname(os.path.abspath(__file__)))
import config
def resample_to_timeframe(df, timeframe):
"""
Resample 1-minute OHLCV data to a higher timeframe.
Parameters
----------
df : pd.DataFrame
1-minute OHLCV data with DatetimeIndex
timeframe : str
Pandas resample frequency: '5min', '15min', '1H', '4H', '1D'
Returns
-------
pd.DataFrame
Resampled OHLCV data
"""
# Ensure index is datetime
if not isinstance(df.index, pd.DatetimeIndex):
df = df.set_index('timestamp')
# Resample OHLCV data
resampled = df.resample(timeframe).agg({
'open': 'first',
'high': 'max',
'low': 'min',
'close': 'last',
'volume': 'sum'
}).dropna()
return resampled
def create_all_timeframes(source_file):
"""
Create datasets for all timeframes from 1-minute source data.
Parameters
----------
source_file : str
Path to 1-minute CSV file
Returns
-------
dict
Dictionary of {timeframe: output_file_path}
"""
print("\n" + "="*70)
print("CREATING MULTI-TIMEFRAME DATASETS")
print("="*70)
# Load source data
print(f"\n📥 Loading source data: {source_file}")
df = pd.read_csv(source_file)
df['timestamp'] = pd.to_datetime(df['timestamp'])
df = df.set_index('timestamp')
print(f" Loaded {len(df):,} 1-minute candles")
print(f" Period: {df.index[0]} to {df.index[-1]}")
# Define timeframes to create
timeframes = {
'5min': '5-minute',
'15min': '15-minute',
'30min': '30-minute',
'1H': '1-hour',
'4H': '4-hour',
'1D': 'daily'
}
output_files = {}
for tf_code, tf_name in timeframes.items():
print(f"\n🔄 Creating {tf_name} dataset...")
# Resample
resampled = resample_to_timeframe(df, tf_code)
# Save
output_file = os.path.join(
config.DATA_DIR,
f"btc_{tf_code.lower()}_binance.csv"
)
# Reset index to have timestamp as column
resampled = resampled.reset_index()
resampled.to_csv(output_file, index=False)
output_files[tf_name] = output_file
print(f" ✓ Created {len(resampled):,} {tf_name} candles")
print(f" ✓ Saved to: {output_file}")
return output_files
def estimate_trades_per_timeframe():
"""Display estimated trades per day for each timeframe."""
print("\n" + "="*70)
print("ESTIMATED TRADING CHARACTERISTICS BY TIMEFRAME")
print("="*70)
print("\nBased on current parameters (MA 15/60, ~2 trades/day on 1-min):\n")
timeframes = [
("1-minute", 1, "~2 trades/day", "High frequency, many signals"),
("5-minute", 5, "~8-12 trades/week", "Active intraday"),
("15-minute", 15, "~2-4 trades/week", "Swing trading"),
("30-minute", 30, "~1-2 trades/week", "Position trading"),
("1-hour", 60, "~5-10 trades/month", "Low frequency"),
("4-hour", 240, "~1-3 trades/month", "Very selective"),
("Daily", 1440, "~2-6 trades/year", "Long-term holds")
]
print(f"{'Timeframe':<15} {'Candle (min)':<15} {'Frequency':<20} {'Style'}")
print("-" * 70)
for tf, mins, freq, style in timeframes:
print(f"{tf:<15} {mins:<15} {freq:<20} {style}")
print("\n💡 Hypothesis:")
print(" - Longer timeframes = Less noise, fewer false signals")
print(" - Better win rate but fewer opportunities")
print(" - May need to adjust MA periods (15/60 → 10/30 or 20/80)")
def main():
"""Main execution."""
print("\n🌍 MULTI-TIMEFRAME DATA CREATOR")
# Use the 2-year Binance data as source
source_file = os.path.join(config.DATA_DIR, "btc_1m_binance_730d.csv")
if not os.path.exists(source_file):
print(f"\n✗ Source file not found: {source_file}")
print(f" Please ensure you have 1-minute data available.")
return
# Create all timeframes
output_files = create_all_timeframes(source_file)
# Show characteristics
estimate_trades_per_timeframe()
print("\n" + "="*70)
print("✅ MULTI-TIMEFRAME DATASETS CREATED")
print("="*70)
print("\n📁 Created files:")
for tf_name, filepath in output_files.items():
print(f" - {tf_name}: {filepath}")
print("\n💡 Next steps:")
print(" 1. Test current parameters on each timeframe:")
print(" - Update config.py to use each timeframe file")
print(" - Run: python run_backtest.py")
print(" - Compare results")
print("\n 2. Optimize parameters per timeframe:")
print(" - Run: python optimize_parameters.py (for each TF)")
print(" - Find best settings for each interval")
print("\n 3. Compare performance:")
print(" - Which timeframe gives best Sharpe ratio?")
print(" - Which has best win rate?")
print(" - Which has lowest drawdown?")
print("="*70 + "\n")
if __name__ == "__main__":
main()
|