Rank-transformation to normality

Description

Rank-transformation to normality of a variable or residuals from GLM analysis.

Usage

rntransform(formula,data,family=gaussian)

Arguments

Details

Rank-transformation to normality generates perfectly normal distribution from ANY distribution, unless many/heavy ties are present in variable (or residuals, if formula is used).

When formula is supplied, this procedure first calls ztransform, and then applies rank transformation to residuals.

Value

Vector containing transformed variable, distributed as standard normal.

Author(s)

Yurii Aulchenko

See Also

ztransform

Examples

	# uniformly distributed variable
	x <- round(runif(200)*100)
	# get 7 missing values
	x[round(runif(7,min=1,max=100))] <- NA
	# Z-transform
	y0 <- ztransform(x)
	# Rank-transform to normality
	y1 <- rntransform(x)
	# test normality of the original and transformed var
	shapiro.test(x)
	shapiro.test(y0)
	shapiro.test(y1)
	# plot histogram
	par(mfcol=c(3,1))
	hist(x)
	hist(y0)
	hist(y1)

Results

R version 3.3.1 (2016-06-21) -- "Bug in Your Hair"
Copyright (C) 2016 The R Foundation for Statistical Computing
Platform: x86_64-pc-linux-gnu (64-bit)

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.

> library(GenABEL)
Loading required package: MASS
Loading required package: GenABEL.data
> png(filename="/home/ddbj/snapshot/RGM3/R_CC/result/GenABEL/rntransform.Rd_%03d_medium.png", width=480, height=480)
> ### Name: rntransform
> ### Title: Rank-transformation to normality
> ### Aliases: rntransform
> ### Keywords: utilities
> 
> ### ** Examples
> 
> 	# uniformly distributed variable
> 	x <- round(runif(200)*100)
> 	# get 7 missing values
> 	x[round(runif(7,min=1,max=100))] <- NA
> 	# Z-transform
> 	y0 <- ztransform(x)
> 	# Rank-transform to normality
> 	y1 <- rntransform(x)
> 	# test normality of the original and transformed var
> 	shapiro.test(x)

	Shapiro-Wilk normality test

data:  x
W = 0.9655, p-value = 0.0001114

> 	shapiro.test(y0)

	Shapiro-Wilk normality test

data:  y0
W = 0.9655, p-value = 0.0001114

> 	shapiro.test(y1)

	Shapiro-Wilk normality test

data:  y1
W = 0.99843, p-value = 0.9997

> 	# plot histogram
> 	par(mfcol=c(3,1))
> 	hist(x)
> 	hist(y0)
> 	hist(y1)
> 
> 
> 
> 
> 
> dev.off()
null device 
          1 
>