Skip to content

"Vibe-Trading: Your Personal Trading Agent"

1.7 KB
Raw Blame
"""Maximum diversification ratio: maximize (w' sigma) / sqrt(w' Sigma w).

``sigma`` is the vector of asset volatilities; ``Sigma`` is the covariance matrix.
Higher DR means more diversification per unit of risk.
"""

from typing import Any, Dict

import numpy as np
import pandas as pd

from backtest.optimizers.base import BaseOptimizer


class MaxDiversificationOptimizer(BaseOptimizer):
    """Maximize diversification ratio (Choueifaty & Coignard)."""

    def _calc_weights(self, ctx: Dict[str, Any]) -> np.ndarray:
        """SLSQP max-DR weights."""
        from scipy.optimize import minimize

        cov = ctx["cov"]
        n = cov.shape[0]
        if n == 0:
            return self._equal_weight(0)

        vols = np.sqrt(np.diag(cov))
        if np.any(vols < 1e-12):
            return self._equal_weight(n)

        def neg_dr(w: np.ndarray) -> float:
            port_vol = np.sqrt(w @ cov @ w)
            if port_vol < 1e-12:
                return 0.0
            return -(w @ vols) / port_vol

        result = minimize(
            neg_dr,
            self._equal_weight(n),
            method="SLSQP",
            bounds=[(0.0, 1.0)] * n,
            constraints={"type": "eq", "fun": lambda w: w.sum() - 1.0},
            options={"maxiter": 200, "ftol": 1e-10},
        )

        if result.success:
            return self._normalize(result.x)
        return self._equal_weight(n)


def optimize(
    ret: pd.DataFrame,
    pos: pd.DataFrame,
    dates: pd.DatetimeIndex,
    lookback: int = 60,
) -> pd.DataFrame:
    """Module-level entry: max-diversification-adjusted positions."""
    return MaxDiversificationOptimizer(lookback=lookback).optimize(ret, pos, dates)