Skip to contents

Normality Tests

Source: R/tests_regular.R
normality_tests.Rd

Set of functions to test the normality of a time series.

Usage

bowmanshenton(data)

doornikhansen(data)

jarquebera(data, k = 0, sample = TRUE)

skewness(data)

kurtosis(data)

Arguments

data

data being tested.

k

number of degrees of freedom to be subtracted if the input time series is a series of residuals.

sample

boolean indicating if unbiased empirical moments should be computed.

Value

A c("JD3_TEST", "JD3") object (see statisticaltest for details).

Functions

  • bowmanshenton(): Bowman-Shenton test

  • doornikhansen(): Doornik-Hansen test

  • jarquebera(): Jarque-Bera test

  • skewness(): Skewness test

  • kurtosis(): Kurtosis test

Examples

x <- rnorm(100) # null
bowmanshenton(x)
#> Value: 0.09858598 
#> P-Value: 0.9519 
doornikhansen(x)
#> Value: 0.05114668 
#> P-Value: 0.9748 
jarquebera(x)
#> Value: 0.04550024 
#> P-Value: 0.9775 
skewness(x)
#> Value: -0.02688914 
#> P-Value: 0.9126 
kurtosis(x)
#> Value: 2.855887 
#> P-Value: 0.7686 

x <- random_t(2, 100) # alternative
bowmanshenton(x)
#> Value: 821.571 
#> P-Value: 0.0000 
doornikhansen(x)
#> Value: 45.82106 
#> P-Value: 0.0000 
jarquebera(x)
#> Value: 908.8287 
#> P-Value: 0.0000 
skewness(x)
#> Value: 2.424057 
#> P-Value: 0.0000 
kurtosis(x)
#> Value: 16.1785 
#> P-Value: 0.0000