
Home

Selforganizology, 2015, 2(4): 65-77
[XML] [EndNote] [RefManager] [BibTex] [

Full PDF (139K)] [Comment/Review Article]

Article

Calculation and statistic test of partial correlation of general correlation measures

WenJun Zhang
School of Life Sciences, Sun Yat-sen University, Guangzhou 510275, China; International Academy of Ecology and Environmental Sciences, Hong Kong

Received 13 July 2015;Accepted 10 August 2015;Published online 1 December 2015
IAEES

Abstract
It is well known that Pearson linear correlations between more than two attributes (nodes, taxa, variables, etc) can be adjusted to partial linear correlations for eliminating indirect between-attribute interactions of other attributes not being tested. In present study I first proposed three correlation measures, revised Dice coefficient, overlap coefficient, and proportion correlation. In addition, I proposed partial correlation measures for some correlation measures, of which Jaccard correlation, revised Dice coefficient, overlap coefficient, and point correlation are for binary attributes; Spearman rank correlation and proportion correlation are for interval value attributes. The full algorithm and Matlab codes (Pearson linear correlation is included also) are given. Users can add other general correlation measures in the Matlab codes.

Keywords partial correlation;correlation measures;proportion correlation;statistic test;data analysis.

International Academy of Ecology and Environmental Sciences. E-mail: office@iaees.org
Copyright © 2009-2024 International Academy of Ecology and Environmental Sciences. All rights reserved.
Web administrator: office@iaees.org, website@iaees.org; Last modified: 2024/5/9

Translate page to: