Summary
We deal with the problem of evaluating the strength of the dependency of a categorical response variable upon a categorical explanatory variable by means of the Goodman and Kruskal’s τ. For given marginals, τ seldom reaches its theoretical upper bound, 1. It seems therefore reasonable to determine the maximum value the index can assume in a class of contingency tables with given marginals. Three kinds of heuristics developed to this aim are presented. An upper bound for the maximum is determined so that it is possible to estimate the relative error of the proposed heuristics.
Similar content being viewed by others
References
Agresti, A. (1990), Categorical Data Analysis, Wiley, New York.
Fréchet, M. (1951), Sur les tableaux de corrélation dont le marges sont données, Ann. de l’Univ. de Lyon, Sect. A.
Kalantari, B., Lari, I., Rizzi, A. & Simeone, B. (1993), Sharp Bounds for the Maximum of the Chi—Square index in a Class of Contingency Tables with Given Marginals, Computational Statistics and Data Analysis, 16, 19–34.
Lauro, N.C. & D’Ambra, L. (1984), L’analyse non symetrique des correspondances, in E. Diday et al. (eds.), Data Analysis and Informatics, III, North Holland.
Piccarreta, R. (1997), L’indice di Goodman e Kruskal e l’analisi non simmetrica delle corrispondenze, Phd. Thesis.
Salvemini, T. (1939), Sugli indici di omofilia, Atti Riunione della Società Italiana di Statistica.
Author information
Authors and Affiliations
Additional information
I wish to thank Prof. Carlo Lauro, for useful conversations and for having stimulated the ideas presented in this paper. I am also grateful to two anonymous referees for valuable comments and suggestions.
Rights and permissions
About this article
Cite this article
Piccarreta, R. Sharp bounds for the maximum of the Goodman-Kruskal τ index in a class of I x J tables with given row and column totals. Computational Statistics 14, 397–418 (1999). https://doi.org/10.1007/PL00022712
Published:
Issue Date:
DOI: https://doi.org/10.1007/PL00022712