linearmodels.iv.gmm.HeteroskedasticWeightMatrix¶

class linearmodels.iv.gmm.HeteroskedasticWeightMatrix(center: bool = False, debiased: bool = False)[source]¶

Heteroskedasticity robust weight estimation

Parameters:¶

center: bool = False¶: Flag indicating whether to center the moment conditions by subtracting the mean before computing the weight matrix.
debiased: bool = False¶: Flag indicating whether to use small-sample adjustments

Notes

The weight matrix estimator is

\[\begin{split}g_i & =z_i\epsilon_i\\ W & =n^{-1}\sum_{i=1}^{n}g'_ig_i\end{split}\]

where \(z_i\) contains both the exogenous regressors and instruments.

Methods

`weight_matrix`(x, z, eps)	param x: Model regressors (exog and endog), (nobs by nvar)

Properties

`config`	Weight estimator configuration