Copyright	(c) Praneya Kumar Alexey Khudyakov 2025
License	BSD-3-Clause
Safe Haskell	None
Language	Haskell2010

Statistics.Test.Levene

Description

Levene's test used to check whether samples have equal variance. Null hypothesis is all samples are from distributions with same variance (homoscedacity). Test is robust to non-normality, and versatile with mean or median centering.

>>> import qualified Data.Vector.Unboxed as VU
>>> import Statistics.Test.Levene
>>> :{
let a = VU.fromList [8.88, 9.12, 9.04, 8.98, 9.00, 9.08, 9.01, 8.85, 9.06, 8.99]
    b = VU.fromList [8.88, 8.95, 9.29, 9.44, 9.15, 9.58, 8.36, 9.18, 8.67, 9.05]
    c = VU.fromList [8.95, 9.12, 8.95, 8.85, 9.03, 8.84, 9.07, 8.98, 8.86, 8.98]
in levenesTest Median [a, b, c]
:}
Right (Test {testSignificance = mkPValue 2.4315059672496814e-3, testStatistics = 7.584952754501659, testDistribution = fDistributionReal 2.0 27.0})

Synopsis

data Center
- = Mean
- | Median
- | Trimmed !Double
levenesTest :: Vector v Double => Center -> [v Double] -> Either String (Test FDistribution)

Documentation

data Center Source #

Center calculation method

Constructors

Mean	Use arithmetic mean
Median	Use median
Trimmed !Double	Trimmed mean with given proportion to cut from each end

Instances

Instances details

Show Center Source #
Instance details Defined in Statistics.Test.Levene Methods showsPrec :: Int -> Center -> ShowS # show :: Center -> String # showList :: [Center] -> ShowS #
Eq Center Source #
Instance details Defined in Statistics.Test.Levene Methods (==) :: Center -> Center -> Bool # (/=) :: Center -> Center -> Bool #

levenesTest Source #

Arguments

:: Vector v Double
=> Center	Centering method
-> [v Double]	Input samples
-> Either String (Test FDistribution)

Main Levene's test function with full error handling