A Measure of Association Based on Gini's Mean Difference

A measure of association between two random variables is proposed, which represents a compromise between the classical correlation coefficient and the rank correlation coefficient. The new measure is based on the covariance between two variables where one variable is taken in its variate values while the other is ranked. The large sample distribution of this measure is studied and some possible applications are mentioned.


Issue Date:
1985
Publication Type:
Working or Discussion Paper
PURL Identifier:
http://purl.umn.edu/232621
Total Pages:
30
Series Statement:
Working Paper
8501




 Record created 2017-04-01, last modified 2017-08-29

Fulltext:
Download fulltext
PDF

Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)