Journal of Statistics: Advances in Theory and ApplicationsAuthor's: Kouji Yamamoto, Mana Aizawa and Sadao Tomizawa
Pages: [17] - [43]
Received Date: July 22, 2016
Submitted by:
DOI: http://dx.doi.org/10.18642/jsata_7100121694
For the analysis of ordinal square contingency tables, Yamamoto et al. [17] considered the sum-symmetry model. The present paper proposes a measure to represent the degree of departure from the sum-symmetry model. The proposed measure is expressed by using the power-divergence or Patil and Taillie’s diversity index. The relationships between the proposed measure and the normal distribution are shown. Some biomedical data are analyzed by using the proposed measure.
diversity index, Kullback-Leibler information, measure, power-divergence, sum-symmetry.
