backtesting-trading-strategies

Backtesting Trading Strategies

Overview

Simulate trading strategies against historical data. Results describe a historical simulation, not expected future performance. The framework includes 8 built-in strategies, the metrics listed below, and parameter optimization.

Key Features:

• 8 pre-built trading strategies (SMA, EMA, RSI, MACD, Bollinger, Breakout, Mean Reversion, Momentum)
• Performance metrics (Sharpe, Sortino, Calmar, VaR, max drawdown)
• Parameter grid search optimization
• Equity curve visualization
• Trade-by-trade analysis

Prerequisites

Before invoking this skill, the user installs the required dependencies in the Python environment used for the simulation:

set -euo pipefail
pip install pandas numpy yfinance matplotlib

Optional for advanced features:

set -euo pipefail
pip install ta-lib scipy scikit-learn

Instructions

1. Fetch historical data (cached to ${CLAUDE_SKILL_DIR}/data/ for reuse):

   复制

   python ${CLAUDE_SKILL_DIR}/scripts/fetch_data.py --symbol BTC-USD --period 2y --interval 1d
   

2. Run a backtest with default or custom parameters:

   复制

   python ${CLAUDE_SKILL_DIR}/scripts/backtest.py --strategy sma_crossover --symbol BTC-USD --period 1y
   python ${CLAUDE_SKILL_DIR}/scripts/backtest.py \
     --strategy rsi_reversal \
     --symbol ETH-USD \
     --period 1y \
     --capital 10000 \
     --params '{"period": 14, "overbought": 70, "oversold": 30}'
   

3. Analyze results saved to ${CLAUDE_SKILL_DIR}/reports/ -- includes *_summary.txt (performance metrics), *_trades.csv (trade log), *_equity.csv (equity curve data), and *_chart.png (visual equity curve).

4. Optimize parameters via grid search to find the best combination:

   复制

   python ${CLAUDE_SKILL_DIR}/scripts/optimize.py \
     --strategy sma_crossover \
     --symbol BTC-USD \
     --period 1y \
     --param-grid '{"fast_period": [10, 20, 30], "slow_period": [50, 100, 200]}'
   

Output

Performance Metrics

Risk Metrics

Trade Statistics

Example Output

Figure-free layout; actual values depend on the input series, fees, slippage, strategy and parameters. Do not present these placeholders as measured returns.

HISTORICAL SIMULATION: [strategy]
[symbol] | [start_date] to [end_date]
Total Return: [computed] | Max Drawdown: [computed]
Sharpe Ratio: [computed] | Win Rate: [computed]
Total Trades: [observed] | Profit Factor: [computed or N/A]

Profit factor is unavailable when no losing trades were observed, including a zero-trade run or a single winning/breakeven trade. A single losing trade yields zero gross profit divided by its loss; this is not evidence of strategy quality. Zero/one-trade samples cannot establish performance, and optimization is an in-sample historical comparison rather than a forecast.

Supported Strategies

Configuration

Create ${CLAUDE_SKILL_DIR}/config/settings.yaml:

data:
  provider: yfinance
  cache_dir: ./data

backtest:
  default_capital: 10000 # Starting simulated capital in quote-currency units
  commission: 0.001 # 0.1% per trade
  slippage: 0.0005 # 0.05% slippage

risk:
  max_position_size: 0.95
  stop_loss: null # Optional fixed stop loss
  take_profit: null # Optional fixed take profit

Error Handling

Use Read to inspect references/errors.md for common issues and solutions.

Examples

See references/examples.md for detailed usage examples including:

• Multi-asset comparison
• Walk-forward analysis
• Parameter optimization workflows

Files

Resources

• yfinance - Yahoo Finance data
• TA-Lib - Technical analysis library
• QuantStats - Portfolio analytics