wald.test {ptmixed}

Wald test for regression coefficients

Description

Compute a multivariate Wald test for one of the following models: Poisson-Tweedie GLMM, negative binomial GLMM, Poisson-Tweedie GLM, negative binomial GLM. The null hypothesis has to be specified in the (matrix) form $L b = k$, where $b$ is the vector of regression coefficients and $L$ and $k$ are defined below

Usage

wald.test(obj, L, k = NULL)

Arguments

obj

an object of class ptglmm (obtained from ptmixed or nbmixed) or ptglm (obtained from ptglm or nbglm)

L

a matrix used to define the hypothesis to test, in the form $L b = k$

k

a vector used to define the hypothesis to test, in the form $L b = k$. Default is a null vector ($L b = 0$)

Value

A data frame with the result of the test

Author(s)

Mirko Signorelli

References

Signorelli, M., Spitali, P., Tsonaka, R. (2021). Poisson-Tweedie mixed-effects model: a flexible approach for the analysis of longitudinal RNA-seq data. Statistical Modelling, 21 (6), 520-545. URL: https://doi.org/10.1177/1471082X20936017

Examples

# generate data
data(df1, package = 'ptmixed')

# estimate one of the following models: a Poisson-Tweedie or 
# negative binomial GLMM (using ptmixed() or nbmixed()), or
# a Poisson-Tweedie or negative binomial GLM (using ptglm() 
# or nbgml())
fit1 = nbglm(formula = y ~ group*time, data = df1)

# define L for beta2 = beta4 = 0
L = matrix(0, nrow = 2, ncol = 4)
L[1, 2] = L[2, 4] = 1
              
# compute multivariate Wald test
wald.test(obj = fit1, L = L, k = NULL)