linearmodels.panel.results.PanelEffectsResults.f_statistic

property PanelEffectsResults.f_statistic : WaldTestStatistic

Joint test of significance for non-constant regressors

Returns:

Statistic value, distribution and p-value

Return type:

linearmodels.shared.hypotheses.WaldTestStatistic

Notes

Classical F-stat that is only correct under an assumption of homoskedasticity. The test statistic is defined as

\[F = \frac{(RSS_R - RSS_U)/ k}{RSS_U / df_U}\]

where \(RSS_R\) is the restricted sum of squares from the model where the coefficients on all exog variables is zero, excluding a constant if one was included. \(RSS_U\) is the unrestricted residual sum of squares. k is the number of non-constant regressors in the model and \(df_U\) is the residual degree of freedom in the unrestricted model. The test has an \(F_{k,df_U}\) distribution.