Premium
Generalization of pair correlation method (PCM) for non‐parametric variable selection
Author(s) -
Héberger Károly,
Rajkó Róbert
Publication year - 2002
Publication title -
journal of chemometrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.47
H-Index - 92
eISSN - 1099-128X
pISSN - 0886-9383
DOI - 10.1002/cem.748
Subject(s) - ranking (information retrieval) , pairwise comparison , contingency table , generalization , statistics , mathematics , selection (genetic algorithm) , variable (mathematics) , correlation , variables , econometrics , artificial intelligence , computer science , mathematical analysis , geometry
The pair correlation method (PCM) has been developed for choosing between two correlated predictor variables (factors) provided that the scatter is caused not only by random effects. The distinction between two variables can be made using an arrangement into a 2 × 2 contingency table. Further on, suitable test statistics can be used to decide the significance of differences between factors. PCM can easily be generalized (GPCM) for variable selection purposes using more than two variables. The comparison of factors can be made pairwise in all possible combinations. If a given statistical test indicates a significant difference between the factors, the following terms are used for the overwhelming and subordinate factors: superior–inferior or winner–loser respectively. Every comparison can mark a factor as superior, inferior or no decision can be made. The following step is ranking of predictor variables. Three ways of ranking have been elaborated: (i) simple ranking, (ii) ranking based on differences and (iii) ranking according to probability‐weighted differences. (Difference here means number of wins minus number of losses.) Suitable examples are presented to show the usefulness and applicability of the method in various conditions. Copyright © 2002 John Wiley & Sons, Ltd.