Resumen
Se estudia un método de estimación paramétrica basado en la minimización del estadísticoD n de Kolmogorov-Smirnov. Se prueba la existencia y unicidad de este estimador en familias de distribuciones monótonas en alguno de sus parámetros y se compara computacionalmente con el método de máxima verosimilitud.
Summary
We study a method of parametric estimation which is based in the minimization ofD n Komogorov-Smirnov’s statistic. It is shown the existence and the unicity of that estimator on families of monotone distributions on some of its parameters. We compare computationally this method with the maximum likelihood one.
This is a preview of subscription content, log in via an institution to check access.
Referencias
BAZARAA, M.S., and SHETTY, C. M.: (1979)Nonlinear Programming. Theory and Algorithms, Wiley.
DANNENBRING, D. G. (1977): «Procedures for Estimating Optimal Solution Values for Large Combinatorial Problems»,Man. Sci., 23, 1273–1283.
PARR, W. C. (1981): «Minimum Distance Estimation. A Bibliography»,Commun. Statist. Theor. Meth., A10 (12), 1205–1224.
ROHATGI, V.K. (1976):An Introduction to Probability Theory and Mathematical Statistic» Wiley.
WOLFE, M.A. (1978):Numerical Methods for Unconstrained Optimization, Van Nostrad Reinhold Company.
WOLFOWITZ, J. (1957): «The Minimum Distance Method»,Annals of Mathematical Statistics, 28, 75–88.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Angel, F.O. Uso del estadisticoD n de Kolmogorov-Smirnov en inferencia parametrica. TDE 3, 177–194 (1988). https://doi.org/10.1007/BF02884333
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF02884333