Quant Buffet API

Metrics API

compute_metrics fields and benchmark-relative stats.

`backtest.metrics.compute_metrics` turns an equity curve into the statistics shown in the lab UI (CAGR, Sharpe, drawdown, etc.).

Usage

metrics = compute_metrics(
    result.equity,
    benchmark=spy_buy_and_hold,  # optional pd.Series, same index as equity
    risk_free=0.0,
    trades_count=len(result.trades),
)
# keys: start, end, years, cagr, volatility, sharpe, sortino,
#       max_drawdown, calmar, daily_win_rate, trades, alpha, beta, ...

Core output fields

Key	Description
`start`, `end`	First and last equity dates (strings).
`years`	Calendar years spanned.
`start_equity`, `end_equity`	Dollar values at start/end.
`total_return`	End/start − 1 (not annualised).
`cagr`	Compound annual growth rate.
`volatility`	Annualised standard deviation of daily returns.
`sharpe`	Excess return / volatility (annualised, `risk_free` subtracted).
`sortino`	Return / downside deviation.
`max_drawdown`	Worst peak-to-trough (negative fraction).
`calmar`	CAGR / \|max_drawdown\|.
`daily_win_rate`	Fraction of positive return days.
`trades`	Trade count passed through from the engine.

Benchmark fields (when benchmark provided)

benchmark_total_return, benchmark_cagr
alpha — annualised Jensen's alpha vs benchmark daily returns
beta — covariance / variance vs benchmark

Note. The lab builds a buy-and-hold benchmark from SPY when available, otherwise the first asset in ASSETS. Display strings (percent formatting) are applied server-side in displayMetrics.