quantlab.portfolio.config

The configs of the portfolio layer.

A portfolio construction rule or covariance estimator is constructed from one of these and exposes it as self.config. They are frozen; a rule’s covariance estimator is declared with component() and written as its own config.

Classes

ActiveRiskConfig	Config of ActiveRiskOptimizer: MeanVarianceConfig and the benchmark risk is measured against.
BlackLittermanConfig	Config of BlackLittermanOptimizer: MeanVarianceConfig, the market and tau.
FactorReturnViewConfig	Config of FactorReturnView: a view on one risk-model factor's return.
FactorRiskStoreEstimatorConfig	Config of FactorRiskStoreEstimator: the factor risk model whose stores it reads.
GrinoldExpectedReturnConfig	Config of GrinoldExpectedReturn: mu = ic * sigma * z.
LedoitWolfEstimatorConfig	Config of LedoitWolfEstimator: a shrunk sample covariance of trailing returns.
MarketReturnViewConfig	Config of MarketReturnView: a view on the market portfolio's return.
MeanVarianceConfig	Config of MeanVarianceOptimizer: Markowitz weights with a turnover penalty.
PredictionExpectedReturnConfig	Config of PredictionExpectedReturn: mu is the prediction as it is; no parameters.
RawExpectedReturnConfig	Config of RawExpectedReturn: mu is the prediction, in return units; no parameters.
RelativeFactorReturnViewConfig	Config of RelativeFactorReturnView: a view on one factor's return less another's.
ReturnViewConfig	The fields every user view of a BlackLittermanOptimizer shares: its value and uncertainty.
TopNConfig	Config of TopNConstructor: equal-weight top-n selection.
TopkDropoutConfig	Config of TopkDropoutConstructor: qlib's TopkDropout.
UnscaledZScoreExpectedReturnConfig	Config of UnscaledZScoreExpectedReturn: mu = z, the z-score itself; no parameters.
ZScoreExpectedReturnConfig	Config of ZScoreExpectedReturn: mu = ic * median(sigma) * z.