Tests and Measures for Homoscedasticity

Provides statistical methods for assessing homoscedasticity (equality of variances) across groups. The package implements classical and robust tests for variance homogeneity, together with supporting measures and utilities for exploratory analysis and hypothesis testing. These tools can be used as diagnostic procedures prior to analyses that assume equal variances.

varequal

An R package for testing whether group variances are equal.

There are also other nice alternatives such as vartest and other friends.

Installation

You can install the package from CRAN with:

install.packages("varequal")

or the development version from GitHub with:

if (!require("pak")) install.packages("pak")
pak::pak("P10911004-NPUST/varequal")

Quick start

is_var_equal(roGFP[[1]], ro ~ grp)

Benchmark

O_O: observe is homo, predict as homo
O_X: observe is homo, predict as hetero
X_X: observe is hetero, predict as hetero
X_O: observe is hetero, predict as homo
Type 1 error: O_X / (O_O + O_X) × 100%
Type 2 error: X_O / (X_X + X_O) × 100%
Accuracy: (O_O + X_X) / (O_O + O_X + X_X + X_O) × 100%

For normally-distributed, moderate sample size (8 ≤ n ≤ 20), and outliers-free data

For normally-distributed, small sample size (3 ≤ n ≤ 7), and outliers-free data.

For normally-distributed and moderate sample size (8 ≤ n ≤ 20) data, with one or two outliers.

TODO

• [x] Ansari-Bradley test
  • Ansari & Bradley, 1960
• [ ] Moses test
  • Moses, 1963
• [ ] Miller test
  • Miller, 1968
• [x] Bartlett's test
• [x] Brown-Forsythe test
• [x] Cochran G test
  • Lam, 2010
• [x] Fligner-Killeen test
  • Fligner & Killeen, 1974
• [ ] Hartley's test
  • Frey, 2010
• [x] Levene's test