pdInd: Constuctor for Positive-Definite Matrix With Zero Covariances...

In gmonette/spida: Collection of miscellaneous functions for mixed models etc. prepared for SPIDA 2009+ (development version)

Description

This function is a constructor for the pdInd class, representing a positive-definite matrix with zero covariances except possibly in the first row and column. If the matrix associated with object is of dimension $n$, it is represented by $n + (n-1)$ unrestricted parameters representing a lower-triangular log-Cholesky decomposition. The first $n$ parameters are the logs of the diagonal elements of the matrix and the last $n-1$ components are the $n-1$ remaining elements of the lower-triangular decomposition corresponding the to the possibly non-zero covariances in the first row.

Usage

pdInd(value = numeric(0), form = NULL, nam = NULL, data = sys.parent(),
  cov = NULL, zero = NULL)

Arguments

value	an option initialization value, which can be any of the following ...
form	an optional one-sided linear formula specifying the row/column names for the matrix represented by object.
nam	and optional vector of character strings specifying the row/column names for the matrix represented by object.
data	and optional data frame i which to evaluate the variables names in value and form. ...
cov	optional position in lower triangle of covariances that are estimated and, thus, possibly non-zero. The default is that the covariances in the first column are estimated and possibly non-zero.
zero	optional way of specifying covariances constrained to be equal to zero. Only the lower triangular portion of the zero is used. The elements that are equal to 0 corresponds to the pattern of elements that are constrained to zero in the covariance matrix.

CAUTION

cov and zero do not work. Until fixed, pdInd only creates the default covariance pattern in which the only non-zero covariances are those with the first element.

gmonette/spida documentation built on May 17, 2019, 7:25 a.m.