mkcov: Generating Covariance for Correlated Data

In gss: General Smoothing Splines

Generating Covariance for Correlated Data

Description

Generate entries of covariance functions for correlated data.

Usage

mkcov.arma(p, q, n)
mkcov.long(id)
mkcov.known(w)

Arguments

p	Order of AR terms.
q	Order of MA terms.
n	Dimension of covariance matrix.
id	Factor of subject ID.
w	Covariance matrix; only the upper triangular part is used.

Details

mkcov.arma generates covariance functions for ARMA(p,q) model.

mkcov.long generates covariance functions for longitudinal data.

mkcov.known allows one to use a known covariance matrix in ssanova9.

Value

A list of three elements.

fun	Covariance matrix to be evaluated through fun(gamma,env) or fun(env).
env	Constants in covariance function.
init	Initial values for correlation parameters.

Note

One may pass list(fun=...,env=...,init=...) directly to the argument cov in calls to ssanova9, or make use of the mkcov.x functions through cov=list("arma",c(p,q)), cov=list("long",id), or cov=list("known",w).

gss documentation built on Aug. 16, 2023, 9:07 a.m.