Journal of Statistics: Advances in Theory and ApplicationsAuthor's: Nobuko Miyamoto, Takashi Kato and Sadao Tomizawa
Pages: [1] - [15]
Received Date: January 22, 2010
Submitted by:
For the analysis of square contingency tables with the same ordinal row and column classifications, Goodman [3] considered the diagonals-parameter symmetry model, which has the structure of null association in the tables constructed by using two diagonals that are equidistant from the main diagonal. The present paper proposes a measure to represent the degree of departure from diagonals-parameter symmetry model. The proposed measure is expressed by using the measure of association between the row and column variables given by Tomizawa et al. [8]. The measure would be useful for comparing the degree of departure from the diagonals-parameter symmetry model in several tables.
association, diagonals-parameter symmetry, power divergence, measure.
