stremo (version 0.2)

is.pd: Do some basic checking to see if a matrix is positive-definite.

Description

A covariance matrix used in structural equation modeling should be positive-definite, as it is going to be inverted during estimation and fit. This function checks whether a Hermitian matrix is positive-definite.

Usage

is.pd(covmat)

Arguments

covmat

A Hermitian matrix.

Value

TRUE or FALSE.

Details

This function checks, in this order:

1. Matrix is Hermitian

2. All covariances are within bounds

3. See if matrix is invertible using solve

4. Check if all eigenvalues are positive

5. Check whether the determinant is positive

See Also

is.within.bounds and is.hermitian.

Examples

data(latta)
is.pd(latta.greenhouse.cov)