Accession Number : ADA186319

Title :   Conditionally Unbiased Bounded Influence Robust Regression with Applications to Generalized Linear Models.

Descriptive Note : Technical rept. Aug 86-Aug 87,

Corporate Author : NORTH CAROLINA UNIV AT CHAPEL HILL DEPT OF STATISTICS

Personal Author(s) : Kunsch, H R ; Stefanski, L A ; Carroll, R J

PDF Url : ADA186319

Report Date : Mar 1987

Pagination or Media Count : 18

Abstract : This document proposes a class of bounded influence robust regression estimators with conditionally unbiased estimating functions given the design. Optimal estimators are found within this class. Applications are made to generalized linear models. An example applying logistic regression to food stamp data is discussed. Keywords: Asymptotic bias; Generalized linear models; Linear regression.

Descriptors :   *ESTIMATES, *LINEAR REGRESSION ANALYSIS, LINEARITY, LOGISTICS, MATHEMATICAL MODELS, BIAS, FOOD STAMPS, VASOCONSTRICTING

Subject Categories : Statistics and Probability

Distribution Statement : APPROVED FOR PUBLIC RELEASE