Title: 估計函數
Finite-Sample Optimal Estimating Functions in the Presence of a Nuisance Parameter
Authors: 張瑋華
Wei Hwa Chang
陳志榮
Chih Rung Chen
統計學研究所
Keywords: finite-sample optimal;estimating function;score;generalized inverse;quasi-score;linear model;non-orthogonal model;finite-sample optimal;estimating function;score;generalized inverse;quasi-score;linear model;non-orthogonal model
Issue Date: 1998
Abstract: In this paper, we first extend some classical criteria for characterizing any finite-sample optimal estimating functions for the parameter of interest in a more general setting. Secondly, we give appropriate sufficient conditions for the existence of any finite-sample optimal estimating function for the parameter of interest in linear or non-orthogonal models. Finally, the example of megalithic stone rings (Angell and Barber, 1977) is discussed thoroughly to illustrate the theory.
In this paper, we first extend some classical criteria for characterizing any finite-sample optimal estimating functions for the parameter of interest in a more general setting. Secondly, we give appropriate sufficient conditions for the existence of any finite-sample optimal estimating function for the parameter of interest in linear or non-orthogonal models. Finally, the example of megalithic stone rings (Angell and Barber, 1977) is discussed thoroughly to illustrate the theory.
URI: http://140.113.39.130/cdrfb3/record/nctu/#NT870337010
http://hdl.handle.net/11536/63999
Appears in Collections:Thesis