feat(backtest): portfolio simulator with configurable strategy and transaction costs
Event-driven simulation: 1-day buy delay, N-day hold, position-size % of cash.
Models entry cost (spread + slippage + commission) and exit cost (spread + commission)
so round-trip is fully parameterised from the CLI.
Reports: annualized return, SPY benchmark, excess return, max drawdown, Sharpe,
per-trade win rate and avg net return.
CLI: python main.py simulate [--holding-days 7] [--spread 0.003] [--slippage 0.002] ...
Also runnable directly: python backtest/simulate.py --help
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Dominik Roth
1467033aa2