M-ESTIMATORS OF THE UNIFORM ASSOCIATION MODEL IN $R\times C$ CONTINGENCY TABLES

Authors : Serpil AKTAŞ, Meral ÇETİN

Pages : 83-88

View : 26 | Download : 10

Publication Date : 2001-12-01

Article Type : Research Paper

Abstract :If all cell counts $n_{ij}$ of a given $R\times C$ contingency table are positive, estimates of the expected frequencies mij can be found by applying any regression estimator to the logarithm of the observed counts. If an $R\times C$ table contains outlierinsert ignore into journalissuearticles values(s);, ordinary least squares estimates will be affected by the outlierinsert ignore into journalissuearticles values(s);. Various authors have proposed several robust estimators sensitive to outliers. In this study, robust estimators were applied to an $R\times C$ contingency table with an outlier to obtain the robust parameter estimates instead of the maximum likelihood insert ignore into journalissuearticles values(ML); estimates, and the results were discussed.
Keywords : Contingency table, Robust regression, Outlier, Uniform association model