covML: Maximum likelihood covariance estimate

In psychonetrics: Structural Equation Modeling and Confirmatory Network Analysis

Description

These functions complement the base R cov function by simplifying obtaining maximum likelihood (ML) covariance estimates (denominator n) instead of unbiased (UB) covariance estimates (denominator n-1). The function covML can be used to obtain ML estimates, the function covUBtoML transforms from UB to ML estimates, and the function covMLtoUB transforms from UB to ML estimates.

Usage

covML(x, ...)
covUBtoML(x, n, ...)
covMLtoUB(x, n, ...)

Arguments

x	A dataset
n	The sample size
...	Arguments sent to the cov function.

Author(s)

Sacha Epskamp <[e-mail]>

Examples

data("StarWars")
Y <- StarWars[,1:10]

# Unbiased estimate:
UB <- cov(Y)

# ML Estimate:
ML <- covML(Y)

# Check:
all(abs(UB - covMLtoUB(ML, nrow(Y))) < sqrt(.Machine$double.eps))
all(abs(ML - covUBtoML(UB, nrow(Y))) < sqrt(.Machine$double.eps))

Example output

Registered S3 methods overwritten by 'huge':
  method    from   
  plot.sim  BDgraph
  print.sim BDgraph
This is psychonetrics 0.8! Note: this is BETA software! Please mind that the package may not be stable and report any bugs! For more information, please see psychonetrics.org, for questions and issues, please see github.com/SachaEpskamp/psychonetrics.

Attaching package: ‘psychonetrics’

The following object is masked from ‘package:graphics’:

    identify

[1] TRUE
[1] TRUE

psychonetrics documentation built on Oct. 26, 2021, 1:06 a.m.