Portfolio Optimization¶

Efficient Frontier¶

from fincore.optimization import efficient_frontier

ef = efficient_frontier(returns, n_points=50)
print(f"Min volatility: {ef['min_variance']['volatility']:.4f}")
print(f"Max Sharpe: {ef['max_sharpe']['sharpe']:.4f}")

Risk Parity¶

from fincore.optimization import risk_parity

rp = risk_parity(returns)
print(f"Weights: {rp['weights']}")

Constrained Optimization¶

from fincore.optimization import optimize

# Max Sharpe ratio
w = optimize(returns, objective="max_sharpe")

# Target return
w = optimize(returns, objective="target_return", target_return=0.15)

# Target risk
w = optimize(returns, objective="target_risk", target_volatility=0.12)

API Reference¶

`fincore.optimization.frontier.efficient_frontier(returns, n_points=50, risk_free_rate=0.0, short_allowed=False, max_weight=1.0)` ¶

Compute the mean-variance efficient frontier.

参数：

名称	类型	描述	默认
`returns`	`DataFrame`	Asset returns (T x N). Columns = asset names.	必需
`n_points`	`int`	Number of points on the frontier.	`50`
`risk_free_rate`	`float`	Annual risk-free rate (used for Sharpe calculation).	`0.0`
`short_allowed`	`bool`	Whether short selling is allowed.	`False`
`max_weight`	`float`	Maximum weight per asset.	`1.0`

返回：

类型	描述
`dict`	'frontier_returns': array of annualised portfolio returns 'frontier_volatilities': array of annualised portfolio volatilities 'frontier_sharpe': array of Sharpe ratios 'frontier_weights': (n_points x N) weight matrix 'min_variance': dict with keys 'weights', 'return', 'volatility' 'max_sharpe': dict with keys 'weights', 'return', 'volatility', 'sharpe' 'asset_names': list of asset names

`fincore.optimization.risk_parity.risk_parity(returns, risk_budget=None, max_iter=1000)` ¶

Compute risk-parity portfolio weights.

Each asset's marginal risk contribution is equalised (or matched to a user-supplied risk budget).

参数：

名称	类型	描述	默认
`returns`	`DataFrame`	Asset returns (T x N).	必需
`risk_budget`	`ndarray`	Target risk budget per asset (sums to 1). Defaults to equal budget `1/N` for each asset.	`None`
`max_iter`	`int`	Maximum solver iterations.	`1000`

返回：

类型	描述
`dict`	'weights': optimal weight array (N,) 'risk_contributions': risk contribution per asset (N,) 'volatility': portfolio annualised volatility 'asset_names': list of asset names

`fincore.optimization.objectives.optimize(returns, objective='max_sharpe', risk_free_rate=0.0, target_return=None, target_volatility=None, short_allowed=False, max_weight=1.0, min_weight=None, sector_constraints=None, sector_map=None)` ¶

Solve a constrained portfolio optimisation problem.

参数：

名称	类型	描述	默认
`returns`	`DataFrame`	Asset returns (T x N).	必需
`objective`	`str`	One of 'max_sharpe', 'min_variance', 'target_return', 'target_risk'.	`'max_sharpe'`
`risk_free_rate`	`float`	Annual risk-free rate.	`0.0`
`target_return`	`float`	Required for `objective='target_return'`. Annualised target.	`None`
`target_volatility`	`float`	Required for `objective='target_risk'`. Annualised target.	`None`
`short_allowed`	`bool`	Allow negative weights.	`False`
`max_weight`	`float`	Upper bound per asset.	`1.0`
`min_weight`	`float`	Lower bound per asset. Defaults to 0 (or `-max_weight` if shorts allowed).	`None`
`sector_constraints`	`dict`	`{sector_name: (min_alloc, max_alloc)}` pairs.	`None`
`sector_map`	`dict`	`{asset_name: sector_name}` mapping (needed with `sector_constraints`).	`None`

返回：

类型	描述
`dict`	'weights': optimal weight array 'return': annualised expected return 'volatility': annualised volatility 'sharpe': Sharpe ratio 'asset_names': list 'objective': objective used