[KSCI] Korea Science Citation Index Service

http://dx.doi.org/10.9708/jksci.2022.27.06.167

Testing the Equality of Several Correlation Coefficients by Permutation Method

Um, Yonghwan (Dept. of Industrial and Management Engineering, Sungkyul University)

Publication Information

Journal of the Korea Society of Computer and Information / v.27, no.6, 2022 , pp. 167-174 More about this Journal

Abstract

In this paper we investigate the permutation test for the equality of correlation coefficients in several independent populations. Permutation test is a non-parametric testing methodology based upon the exchangeability of observations. Exchangeability is a generalization of the concept of independent, identically distributed random variables. Using permutation method, we may construct asymptotically exact test. This method is asymptotically as powerful as standard parametric tests and is a valuable tool when the sample sizes are small and normality assumption cannot be met. We first review existing parametric approaches to test the equality of correlation coefficients and compare them with the permutation test. At the end, all the approaches are illustrated using Iris data example.

Keywords

permutation test; equality of correlation coefficients; non-parametric testing; exchangeability; normality assumption;

Citations & Related Records

Reference

1	Y. Um, Testing the equality of two linear regression models: Comparison between Chow test and a permutation test, Journal of The Korea Society of Computer and Information, 26, 8, 157-164, 2021. DOI
2	A. Donner and B. Rosner, On inference concerning a common correlation coefficient, Applied Statistics, 29, 69-76, 1980. DOI
3	A. A. Jafari and M. R. Kazemi, Computational approach test for inference about several correlation coefficients: equality and common, Communications in Statistics - Simulation and Computation, 46, 3, 2043-2056, 2017. DOI
4	Y. Um, Permutation est for the equality of several independent Cronbach's alpha coefficients, Journal of The Korea Society of Computer and Information, 24, 6, 159-164, 2019. DOI
5	J. Neyman, Optimal asymptotic tests of composite statistical hypothesies, In: Grenander, V. Q., ed. Probability and Statistics The Harold Cramer Volume. New York: Wiley, p213-234, 1959.
6	R. A. Fisher, The Design of Exprriments, 8th Ed.: Oliver & Boyd: Edinburgh, 1966.
7	K. Pearson, On a method of determining whether a sample of size n supposed to have been drawn from a parent population having a known probability integral has probably been drawn at random, Biometrika, 25, 379-410, 1933. DOI
8	F. N. David, Tables of ordinates and probability Integral of the distribution of the correlation coefficients in small samples, London, Cambridge University Press, 1938.
9	K. Krishnamoorthy and Y. Xia, Inferences on correlation coefficients: One-sample independent and correlated cases, Journal of Statistical Planning and Inference, 137, 7, 2362-2379, 2007. DOI
10	I. Olkin and J. D. Finn, Correlation redux, Psychological Bulletin, 118, 155-164, 1995. DOI
11	H. C. Kraemer, On estimation and hypothesis testing problems for correlation coefficients, Psychometrika, 40, 4, 473-485, 1975. DOI
12	S. R. Paul, Estimation of and testing significance for a common correlation coefficient, Communications in Statistics - Theory and Methods, 17, 1, 39-53, 1988. DOI
13	S. R. Paul, Test for the equality of several correlation coefficients, The Canadian Journak of Statistics, 17, 2, 217-227, 1989. DOI
14	Y. Um, Combining independent permutation p-values associated with multi-sample location test data, Journal of The Korea Society of Computer and Information, 25, 7, 175-182, 2020. DOI
15	Y. Um, Permutation analysis of split-half reliability coefficient, 22, 7, 133-139, 2017. DOI
16	R. A. Fisher, A design of experiment, Oliver & Boyd, Edinburgh, 1935.
17	P. W. Mielke and K. J. Berry, Permutation methods : A distance function approach, Springer-Verlag, New York. 2001.
18	X. Liu, S. Liu and C. X. Ma, Testing equality of correlation coefficients for paired binary data from multiple groups, Journal of Statistical Computation and Simulation, 86, 9, 1686-1696, 2015.