linearmodels.asset_pricing.covariance.HeteroskedasticCovariance

class HeteroskedasticCovariance(xe, *, jacobian=None, inv_jacobian=None, center=True, debiased=False, df=0)[source]

Heteroskedasticity robust covariance estimator

Parameters
xendarray

Scores/moment conditions

jacobianndarray

Jacobian. One and only one of jacobian and inv_jacobian must be provided

inv_jacobian{ndarray, None}

Inverse jacobian. One and only one of jacobian and inv_jacobian must be provided. Default is None.

centerbool

Flag indicating to center the scores when computing the covariance. Default is true.

debiasedbool

Flag indicating to use a debiased estimator. Default is False.

dfint

Degree of freedom value ot use if debiasing. Default is 0.

Attributes
config
cov

Compute parameter covariance

inv_jacobian

Inverse Jacobian

jacobian

The Jacobian

s

Score/moment condition covariance

square

Flag indicating if jacobian is square

Methods

Properties

config

cov

Compute parameter covariance

inv_jacobian

Inverse Jacobian

jacobian

The Jacobian

s

Score/moment condition covariance

square

Flag indicating if jacobian is square