backtesting

Validate backtest engines by verifying outputs against expected metrics like Sharpe, Sortino, drawdown, and CAGR. Ensure reproducible results and maintain test coverage, updating scripts when interfaces change.

Walk-forward backtesting evaluates a strategy sequentially over rolling time windows, while vectorized backtesting applies calculations across entire datasets simultaneously. Both approaches require explicit metric validation and cost modeling for accurate results.

Model slippage and transaction costs by integrating cost adjustments into your backtest evaluation flow. Validate cost modeling alongside equity and position accounting to ensure performance metrics accurately reflect real trading conditions.

Report key performance metrics including Sharpe ratio, Sortino ratio, drawdown, and CAGR when testing trading strategies. Explicit validation of these metrics ensures reliable assessment of strategy viability.

Yes, when interfaces change, script updates under app/scripts/ are required to maintain backtest engine functionality. This ensures continued reproducible outputs and test coverage across vectorized and walk-forward workflows.

Vectorized backtesting applies calculations across entire datasets simultaneously, making it suitable for efficient large-scale strategy evaluation. Validate results against expected metrics and ensure cost modeling accounts for slippage impacts.