z-logo
open-access-imgOpen Access
REGIONAL DEPENDENCE FOR CONTINUOUS BIVARIATE DENSITIES
Author(s) -
Holland Paul W.,
Wang Yuchung J.
Publication year - 1986
Publication title -
ets research report series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.235
H-Index - 5
ISSN - 2330-8516
DOI - 10.1002/j.2330-8516.1986.tb00170.x
Subject(s) - bivariate analysis , contingency table , independence (probability theory) , joint probability distribution , cartesian product , mathematics , statistics , contingency , econometrics , set (abstract data type) , measure (data warehouse) , product (mathematics) , distribution (mathematics) , mathematical analysis , computer science , combinatorics , geometry , data mining , philosophy , linguistics , programming language
When the region of support for a continuous bivariate density is not a cartesian product set, the joint distribution cannot be independent. This is closely related to the effects caused by “structural zeros” in two‐way contingency tables. Nonrectangular regions of support for continuous densities are analogous to the incomplete two‐way tables. Quasi‐independence is introduced to replace independence. A measure of association to quantify the degree of dependence due to region is defined, discussed, and applied to several examples.

The content you want is available to Zendy users.

Already have an account? Click here to sign in.
Having issues? You can contact us here