Double linear diagonals-parameter symmetry and decomposition of double symmetry for square tables

作者：Tahata Kouji^*; Tomizawa Sadao

来源：Statistical Methods and Applications, 2010, 19(3): 307-318.

DOI：10.1007/s10260-009-0127-y

• Bivariate normal distribution
• Mean equality
• Orthogonal decomposition
• Point-symmetry
• Symmetry

For square contingency tables with ordered category, the present paper proposes the double linear diagonals-parameter symmetry (D-LDPS) model which implies the structure of both asymmetry with respect to the main diagonal and with respect to the reverse diagonal in the table. The D-LDPS model may be appropriate for a square ordinal table if it is reasonable to assume an underlying bivariate normal distribution with equal marginal variances. The present paper also gives the orthogonal decomposition of the double symmetry model into the D-LDPS model and the double marginal mean equality model. An example is given.

• 出版日期2010-8