mc_manova_disp: MANOVA-Type Test for Dispersion Components of mcglm Models

In mcglm: Multivariate Covariance Generalized Linear Models

MANOVA-Type Test for Dispersion Components of mcglm Models

Description

Performs a MANOVA-type Wald test for the dispersion parameters of multivariate covariance generalized linear models fitted using mcglm. The test is based on quadratic forms of the estimated dispersion parameters and their covariance matrix, yielding statistics analogous to the Hotelling–Lawley trace.

Usage

mc_manova_disp(object, idx, effect_names, ...)

Arguments

object	An object of class "mcglm".
idx	An integer vector defining the grouping structure of dispersion parameters to be tested.
effect_names	A character vector with labels for the tested dispersion effects.
...	Further arguments (currently not used).

Value

A data frame containing the MANOVA-type test results for the dispersion parameters with the following columns:

Effects

Names of the tested dispersion effects.

Df

Degrees of freedom associated with each effect.

Hotelling-Lawley

Hotelling–Lawley trace statistic.

Chi-square

Chi-square test statistic.

p-value

P-values from the chi-square approximation.

Author(s)

Wagner Hugo Bonat, [e-mail]

See Also

mcglm, mc_manova, coef.mcglm, vcov.mcglm

mcglm documentation built on Jan. 9, 2026, 1:07 a.m.