/rcdm

Primary LanguageR

rcdm

Random Covariance Dynamic Model

Description

This R package implements a random covariance dynamic model (RCDM) for joint estimation of sparse time-varying precision matrices.

Overview of main functions

Function Description
randDynCov Implements the Random Covariance Dynamic Model (RCDM) for joint estimation of sparse time-varying precision matrices. Optimization is conducted using block coordinate descent.
randDynCov_cluster Implements the Random Covariance Dynamic Model (RCDM) for joint estimation of sparse time-varying precision matrices with pre-defined clusters(knots).
bic_cal_m Calculates the modified BIC for the RCDM.

Installation

Install the latest version of the package from GitHub with the following R code:

if("devtools" %in% installed.packages() == FALSE) {
    install.packages("devtools")
}

devtools::install_github("Jackie-Zhu17/rcdm")

Example

For a detailed simulation example for implementing the Random Covariance Dynamic Model (RCDM), see this.