Skip to content

GLMMs with Adaptive Gaussian Quadrature - GLMMadaptive 0.5-1

1 message · Dimitris Rizopoulos

#
Dear R mixed-model users,

A new version of GLMMadaptive (0.5-1) has been rolled out on CRAN.

Summary: GLMMadaptive fits mixed effects models using adaptive
Gaussian quadrature to approximate the integrals over the random
effects, allowing also for user-specified models.

Website: https://drizopoulos.github.io/GLMMadaptive/

New features:

- Support is provided for the **effects** and **ggeffects** packages for 
producing effect plots.

- Support is provided for the **DHARMa** package for checking the 
goodness-of-fit of fitted mixed models using scaled simulated residuals. 
Examples can be found in: vignette("Goodness_of_Fit", package = 
"GLMMadaptive")

- Function marginal_coefs() for calculating coefficients with a marginal 
/ population-averaged interpretation has a faster implementation.

- The optimizer nlminb() can now also be invoked using the new control 
argument 'optimizer'; a new vignette describes how to control the 
optimization and numerical integration procedures in the package: 
vignette("Optimization", package = "GLMMadaptive")

- New family object students.t() for fitting (robust) linear mixed 
models with a Student's t distribution for the error terms.

As always, any kind of feedback is more than welcome.

Best,
Dimitris