statsmodels.genmod.families.family.Family.resid_dev¶

Family.resid_dev(endog, mu, var_weights=1.0, scale=1.0)[source]¶

The deviance residuals

Parameters:¶

endogarray_like

The endogenous response variable

muarray_like

The inverse of the link function at the linear predicted values.

var_weightsarray_like

1d array of variance (analytic) weights. The default is 1.

scalefloat, optional

An optional scale argument. The default is 1.

Returns:¶

resid_devfloat

Deviance residuals as defined below.

Notes

The deviance residuals are defined by the contribution D_i of observation i to the deviance as

\[resid\_dev_i = sign(y_i-\mu_i) \sqrt{D_i}\]

D_i is calculated from the _resid_dev method in each family. Distribution-specific documentation of the calculation is available there.