Premium
The Ordered Weighted Average in the Variance and the Covariance
Author(s) -
Merigó José M.,
Guillén Montserrat,
Sarabia José M.
Publication year - 2015
Publication title -
international journal of intelligent systems
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.291
H-Index - 87
eISSN - 1098-111X
pISSN - 0884-8173
DOI - 10.1002/int.21716
Subject(s) - covariance , parameterized complexity , variance (accounting) , mathematics , weighted arithmetic mean , operator (biology) , covariance intersection , interval (graph theory) , algorithm , statistics , computer science , combinatorics , estimation of covariance matrices , biochemistry , chemistry , accounting , repressor , transcription factor , business , gene
The ordered weighted average (OWA) is an aggregation operator that provides a parameterized family of aggregation operators between the minimum and the maximum. This paper analyzes the use of the OWA in the variance and the covariance. It presents several extensions by using a unified framework between the weighted average and the OWA. Furthermore, it also develops other generalizations with induced aggregation operators and by using quasi‐arithmetic means. Several measures of correlation by using the OWA are introduced including a new type of Pearson coefficient. The paper ends with some numerical examples focused on the construction of interval and fuzzy numbers with the variance and the covariance.