ops-package: Optimal Power Space Transformation

In ops: Optimal Power Space Transformation

Description

Comparison of data by Pearson product-moment correlation coefficients is prone to outliers. The problem can be alleviated by normalizing data with outliers before computing the Pearson correlation coefficient. The sample provides such normalization by optimal power space transformation.

Details

Package:	ops
Type:	Package
Version:	1.0
Date:	2012-02-12
License:	Apache License 2.0
LazyLoad:	yes

Use function findP() to compute the exponent for an optimal power space transformation. Optionally, pre-filter the dataset excluding values that are equal or less a certain threshold by calling function filter().

Author(s)

Micha Sammeth <[e-mail]>

References

Ribeca P. and Sammeth M. (under review)

Examples

x=cbind(rexp(1000),rexp(1000))
p=findP(x)$maxIQR
y=x^p

Example output

Attaching package: 'ops'

The following object is masked from 'package:stats':

    filter

ops documentation built on May 1, 2019, 7:55 p.m.