Covariate-Varying Networks (CVN)

CVN is an R package for estimating high-dimensional Gaussian graphical models that change with multiple external covariates. The model is flexible, in the sense that complex smoothing patterns between the individual graphs can be used.

See the vignette for more information.

?CVN

Installation

To install, simply type in R

devtools::install_github("bips-hb/CVN")

or

install.packages('CVN', repos = c('https://bips-hb.r-universe.dev', 'https://cloud.r-project.org'))

Reference

Louis Dijkstra, Arne Godt, Ronja Foraita (2024).
Inferring High-Dimensional Dynamic Networks Changing with Multiple Covariates
https://arxiv.org/abs/2407.19978

Acknowledgment

DFG - Deutsche Forschungsgemeinschaft (FO 1045/2-1)

Author Team

Louis Dijkstra (Creator and main author)
Ronja Foraita (Maintainer, author)
Lukas Burk (Contributor)

Contact

Ronja Foraita
Leibniz Institute for Prevention Research and Epidemiology - BIPS
E-Mail: R (at) leibniz-bips.de