TuneExplainedVarianceCov: Tuning function (covariance)

In fake: Flexible Data Simulation Using the Multivariate Normal Distribution

Tuning function (covariance)

Description

Computes the difference in absolute value between the desired and observed proportion of explained variance from the first Principal Component of a Principal Component Analysis applied on the covariance matrix. The precision matrix is obtained by adding u to the diagonal of a positive semidefinite matrix. This function is used to find the value of the constant u that generates a covariance matrix with desired proportion of explained variance.

Usage

TuneExplainedVarianceCov(u, ev_xx = NULL, lambda)

Arguments

u	constant u added to the diagonal of the precision matrix.
ev_xx	desired proportion of explained variance. If ev_xx=NULL, the obtained proportion of explained variance is returned.
lambda	eigenvalues of the positive semidefinite precision matrix.

Value

The absolute difference in proportion of explained variance (if ev_xx is provided) or observed proportion of explained variance (if ev_xx=NULL).

fake documentation built on April 14, 2023, 12:37 a.m.