linearmodels.iv.absorbing.AbsorbingLSResults.f_statistic¶

property AbsorbingLSResults.f_statistic : WaldTestStatistic¶

Model F-statistic.

Returns:¶

Test statistic for null all coefficients excluding constant terms are zero.

Return type:¶

linearmodels.shared.hypotheses.WaldTestStatistic

Notes

Despite name, always implemented using a quadratic-form test based on estimated parameter covariance. Default is to use a chi2 distribution to compute p-values. If debiased is True, divides statistic by number of parameters tested and uses an F-distribution.

This version of the F-statistic directly uses the model covariance estimator and so is robust against the same specification issues.

An InvalidTestStatistic is returned if the model has no non-constant exogenous terms, or if the covariance of the parameters is singular, e.g., the dependent variable is all zero, so that the model fits exactly.