mvrmat: Design Matrices for Modeling Elements of Regressograms

In priyakohli5/MLGM: Multivariate Longitudinal Graphical Models

Description

mvrmat returns design matrices for polynomial models used for modeling elements of modified Cholesky block decomposition.

Usage

mvrmat(time,j,r,s)

Arguments

time	a vector with T equally or unequally spaced time points.
j	a positive integer for the number of outcomes.
r	a positive integer for the order of time-lag polynomial in constructing design matrix for regressogram.
s	a positive integer for the order of time polynomial in constructing design matrix for innovariogram.

Value

design matrices for models selected for regressogram and innovariogram. Specifically,

• U is the design matrix for time-lag polynomials selected for regressogram.

• V is the design matrix for time polynomials selected for log innovariogram.

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)
time <- c(0, 2, 4, 6, 8, 18, 24, 32, 48, 72)
t <- length(time)
j <- 4
n <- 44
r <- 3
s <- 2
design_mat <- mvrmat(time,j,r,s)

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