Correctness of Kruskal's Algorithms for Monotone Regression with Ties | Psychometrika | Cambridge Core (original) (raw)
Abstract
Kruskal has proposed two modifications of monotone regression that can be applied if there are ties in nonmetric scaling data. In this note we prove Kruskal's conjecture that his algorithms give the optimal least squares solution of these modified monotone regression problems. We also propose another (third) approach for dealing with ties.
References
Van Eeden, C. Testing and estimating ordered parameters of probability distributions. Unpublished doctoral dissertation, University of Amsterdam, 1958.Google Scholar
Barlow, R. E., Bartholomew, D. J., Bremner, J. M., & Brunk, H. D. Statistical inference under order restrictions, 1972, New York: Wiley.Google Scholar
Bradley, R. A., Katti, S. K., & Coons, I. J. Optimal scaling for ordered categories. Psychometrika, 1962, 27, 355–374.CrossRefGoogle Scholar
Hayashi, C. Minimum dimension analysis MDA: One of the methods of multidimensional quantification. Behaviormetrika, 1974, 1, 1–24.CrossRefGoogle Scholar
Kruskal, J. B. Nonmetric multidimensional scaling: A numerical method. Psychometrika, 1964, 29, 115–129.CrossRefGoogle Scholar