Learn R Programming

hmgm (version 1.0.3)

High-Dimensional Mixed Graphical Models Estimation

Description

Provides weighted lasso framework for high-dimensional mixed data graph estimation. In the graph estimation stage, the graph structure is estimated by maximizing the conditional likelihood of one variable given the rest. We focus on the conditional loglikelihood of each variable and fit separate regressions to estimate the parameters, much in the spirit of the neighborhood selection approach proposed by Meinshausen-Buhlmann for the Gaussian Graphical Model and by Ravikumar for the Ising Model. Currently, the discrete variables can only take two values. In the future, method for general discrete data and for visualizing the estimated graph will be added. For more details, see the linked paper.

Copy Link

Version

Install

install.packages('hmgm')

Monthly Downloads

46

Version

1.0.3

License

GPL (>= 2)

Maintainer

Mingyu Qi

Last Published

October 7th, 2020

Functions in hmgm (1.0.3)

pargen

Generating parameters according to the graph
hmgm

High-dimensional Mixed Graphical Models Estimation
pargroup

Function to partition overlapping groups into non-overlapping groups
fitadj

Obtain the adjascent matrix by thresholding the adj norm matrix
hmgm-package

High-dimensional mixed graphical models estimation
datagen

Data generator
edgenorm

Calculate the group L2 norm for each pair of edges