mvchol: Multivariate Cholesky Block Decomposition

In priyakohli5/MLGM: Multivariate Longitudinal Graphical Models

Description

mvchol returns modified Cholesky block decomposition (MCBD) of the sample covariance matrix for multivariate longitudinal data with j outcomes.

Usage

mvchol(S,t,j)

Arguments

S	Sample covariance matrix of the multivariate longitudinal data.
t	positive integer indicating the number of repeated measurements.
j	positive integer representing the number of outcomes.

Value

Elements from MCBD of the covariance matrix:

• T.mat is the regression coefificients matrix T matrix in the MCBD.

• Phit is Φ_{t} matrix with regression coefficients stacked as φ_{t1}, φ_{t2}, ..., φ_{tk}.

• bigPhi is Φ matrix with regression coefficients stacked as Φ_{2}, Φ_{3}, ..., Φ_{T}.

• D is the innovation variance matrix with block diagonal elements D_1, D_2, … , D_t.

• Dt are the j*j innovation variance matrices.

References

Kohli, P. Garcia, T. and Pourahmadi, M. 2016 Modeling the Cholesky Factors of Covariance Matrices of Multivariate Longitudinal Data, Journal of Multivariate Analysis, 145, 87-100.

Examples

data(Tcells)
S <- cov(Tcells)
t <- 10
j <- 4
MCBD <- mvchol(S,t,j)
names(MCBD)

priyakohli5/MLGM documentation built on April 24, 2021, 4:22 p.m.