data.scale: Scale a matrix

In mht: Multiple Hypothesis Testing for Variable Selection in High-Dimensional Linear Models

Description

Scale the data so each column has mean 0 and variance 1. This function is used as a pre-processing step to prep the data for analysis in all functions of the mht package.

Usage

data.scale(data,warning)

Arguments

data	Input matrix of dimension n * p; each row is an observation vector. The intercept should be included in the first column as (1,...,1). If not, it is added.
warning	Logical value. A warning message is printed if the intercept is added. Default is TRUE.

Details

Scale the data so each column has mean 0 and variance 1. If we note x a column of the output scaled matrix -except the first one which is the intercept, we have sum(x)=0 and sum(x^2)/n=1.

Value

data	Scaled data.
intercept	Logical value. TRUE if the intercept was already included in the input data.
means.data	Vector of means of the input data matrix.
sigma.data	Vector of variances of the input data matrix.

References

Multiple hypotheses testing for variable selection; F. Rohart 2011

Examples

## Not run: 
x=matrix(rnorm(100*20),100,20)
res=data.scale(x)
x.scaled=res$data
means.x=res$means.data
sigma.x=res$sigma.data

## End(Not run)

mht documentation built on May 2, 2019, 11:49 a.m.