bothsidesmodel.chisquare: Test subsets of beta are zero

In msos: Data Sets and Functions Used in Multivariate Statistics: Old School by John Marden

Description

Tests the null hypothesis that an arbitrary subset of the β _{ij}'s is zero, based on the least squares estimates, using the χ^2 test as in Section 7.1. The null and alternative are specified by pattern matrices P_0 and P_A, respectively. If the P_A is omitted, then the alternative will be taken to be the unrestricted model.

Usage

bothsidesmodel.chisquare(
  x,
  y,
  z,
  pattern0,
  patternA = matrix(1, nrow = ncol(x), ncol = ncol(z))
)

Arguments

x	An N x P design matrix.
y	The N x Q matrix of observations.
z	A Q x L design matrix.
pattern0	An N x P matrix of 0's and 1's specifying the null hypothesis.
patternA	An optional N x P matrix of 0's and 1's specifying the alternative hypothesis.

Value

A 'list' with the following components:

Theta

The vector of estimated parameters of interest.

Covtheta

The estimated covariance matrix of the estimated parameter vector.

df

The degrees of freedom in the test.

chisq

T^2 statistic in (7.4).

pvalue

The p-value for the test.

See Also

bothsidesmodel, bothsidesmodel.df, bothsidesmodel.hotelling, bothsidesmodel.lrt, and bothsidesmodel.mle.

Examples

# TBA - Submit a PR!

Example output

Loading required package: mclust
Package 'mclust' version 5.4.3
Type 'citation("mclust")' for citing this R package in publications.
Loading required package: tree

msos documentation built on Oct. 31, 2020, 9:07 a.m.