Abstract
The methods of optimal scaling are usually formulated as the maximization problem of a ratio of quadratic forms, and the optimal scores are obtained by solving an eigenequation. However, there sometimes exist order relations among categories. For such cases, Bradley, Katti and Coons [2] proposed an algorithm to maximize the criterion under complete order restrictions. Nishisato and Arri [7] discussed the case of partial order and proposed an algorithm using separable programming. Their method is, however, limited to a special type of partial order. Avoiding this limitation, the present paper gives a generalized formulation applicable to arbitrary order restrictions and proposes an efficient algorithm using Wolfe's reduced gradient method. Numerical examples are provided to show the validity, the rapidness of convergence and the stability of the procedure.
Similar content being viewed by others
References
Biggs, N. (1974).Algebraic Graph Theory, Camridge Univ. Press.
Bradley, R. A., Katti, S. K. and Coons, I. J. (1962). Optimal scaling for ordered categories,Psychometrika,27, 355–374.
Busacker, R. G. and Saaty, T. L. (1965).Finite Graphs and Networks: An Introduction and Applications, McGraw-Hill.
Fisher, R. A. (1958).Statistical Methods for Research Workers, 13th ed. Oliver & Boyd.
Hayashi, C. (1952). On the prediction of phenomena from qualitative data and the quantification of qualitative data from the mathematico-statistical point of view,Ann. Inst. Statist. Math.,3, 69–98.
Jacobi, S. L. S., Kowalik, J. S. and Pizzo, J. T. (1972).Iterative Methods for Nonlinear Optimization Problems, Prentice-Hall.
Nishisato, S. and Arri, P. S. (1975). Nonlinear programming approach to optimal scaling of partially ordered categories,Psychometrika,40, 525–548.
Tanaka, Y. (1972). On quantification theory with ordered categorical outside variable,Standardization and Quality Control,25, 31–39 (in Japanese).
Tanaka, Y. (1978). Some generalized methods of optimal scaling and their asymptotic theories: The case of multiple responses-multiple factors,Ann. Inst. Statist. Math.,30, A, 329–348.
Tanaka, Y. (1979). Some generalized methods of optimal scaling and their asymptotic theories: The case of single response-multiple factors, to appear.
Wolfe, P. (1963). Methods of Nonlinear Programming,Recent Advances in Mathematical Programming R. L. Graves and P. Wolfe (eds.), McGraw-Hill.
Additional information
Takeda Chemical Industries, Ltd.
Now at Okayama University.
About this article
Cite this article
Tanaka, Y. Optimal scaling for arbitrarily ordered categories. Ann Inst Stat Math 31, 115–124 (1979). https://doi.org/10.1007/BF02480269
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/BF02480269