On asymptotically efficient recursive estimation of the
Teoriâ veroâtnostej i ee primeneniâ, Tome 25 (1980) no. 3, pp. 577-587
Citer cet article
Voir la notice de l'article provenant de la source Math-Net.Ru
The estimates of the location parameter on the base of independent observations $Y_1,\dots,Y_n$ with common distribution having density $p(y)$ are considered. We construct a recursive procedure which is uniformly asymptotically (as $n\to\infty$) efficient in the strong sense when the following conditions are fulfilled: (a) $p(y)$ is absolutely continuous, (b) Fisher's information $I(p)=\int(p')^2p^{-1}\,dy$ is finite.