Abstract
Consider the problems of the continuous invariant estimation of a distribution function with a wide class of loss functions. It has been conjectured for long that the best invariant estimator is minimax for all sample sizes n≥1. This conjecture is proved in this short note.
Similar content being viewed by others
References
Aggarwal, O. P. (1955). Some minimax invariant procedures of estimating a cumulative distribution function, Ann. Math. Statist., 26, 450–462.
Brown, L. D. (1988). Admissibility in discrete and continuous invariant nonparametric problems and in their multivariate analogs, Ann. Statist., 16, 1567–1593.
Cohen, M. P. and Kuo, L. (1985). The admissibility of the empirical distribution function, Ann. Statist., 13, 262–271.
Dvoretzky, A., Kiefer, J. and Wolfowitz, J. (1956). Asymptotic minimax character of the sample distribution function and of the classical multinomial estimator, Ann. Math. Statist., 27, 642–669.
Ferguson, T. S. (1967). Mathematical Statistics, a Decision Theoretic Approach, p. 197, Academic Press, New York.
Yu, Q. (1988). Inadmissibility of the best invariant estimator of a distribution function, Sankhyā Ser. A (to appear).
Yu, Q. (1989a). Inadmissibility of the empirical distribution function in continuous invariant problems, Ann. Statist., 17, 1347–1359.
Yu, Q. (1989b). Admissibility of the best invariant estimator of a distribution function, Statist. Decisions, 7, 1–14.
Yu, Q. (1989c). Methodology for the invariant estimation of a continuous distribution function, Ann. Inst. Statist. Math., 41, 503–520.
Yu, Q. (1989d). Admissibility of the empirical distribution function in the invariant problem, Statis. Decisions, 7, 383–398.
Yu, Q. and Chow, M. S. (1991). Minimaxity of the empirical distribution function in invariant problem, Ann. Statist., 19, 935–951.
Author information
Authors and Affiliations
Additional information
Partially supported by National Science Foundation Grant DMS 9001194.
About this article
Cite this article
Yu, Q. Minimax invariant estimator of a continuous distribution function. Ann Inst Stat Math 44, 729–735 (1992). https://doi.org/10.1007/BF00053401
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF00053401