Ranges of Measures of Association for Familial Binary Variables |
| |
Authors: | N. Rao Chaganty Yihao Deng |
| |
Affiliation: | 1. Department of Mathematics and Statistics , Old Dominion University , Norfolk, Virginia, USA rchagant@odu.edu;3. Department of Mathematical Sciences , Indiana University Purdue University–Fort Wayne , Fort Wayne, Indiana, USA |
| |
Abstract: | Familial binary data occur in a wide range of scientific investigations. Numerous measures of association have been proposed in the literature for the study of intra-family dependence of the binary variables. These measures include correlations, odd ratios, kappa statistics, and relative risks. We study the permissible ranges of these measures of association such that a joint distribution exists for the familial binary variables. Our results are useful for developing efficient estimation methods for the measures of association. |
| |
Keywords: | Familial correlations GEE Kappa statistics Odds ratios Relative risks |
|
|