ASSESSING CASE INFLUENCE ON CONFIDENCE-INTERVALS IN NONLINEAR-REGRESSION

被引：3

作者：

DZIECIOLOWSKI, K ^{[1
]}

ROSS, WH ^{[1
]}

机构：

[1] QUEENS UNIV,KINGSTON K7L 3N6,ONTARIO,CANADA

来源：

CANADIAN JOURNAL OF STATISTICS-REVUE CANADIENNE DE STATISTIQUE | 1990年 / 18卷 / 02期

关键词：

casedeletion model; Curvature; influence; likelihood distance; nonlinear regression; profile t‐plot;

D O I：

10.2307/3315561

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

Diagnostic techniques are proposed for assessing the influence of individual cases on confidence intervals in nonlinear regression. The technique proposed uses the method of profile t‐plots applied to the case‐deletion model. The effect of the geometry of the statistical model on the influence measures is assessed, and an algorithm for computing case‐deleted confidence intervals is described. This algorithm provides a direct method for constructing a simple diagnostic measure based on the ratio of the lengths of confidence intervals. The generalization of these methods to multiresponse models is discussed. Copyright © 1990 Statistical Society of Canada

引用

页码：127 / 139

页数：13

共 24 条

[1]

Amari S., 1985, LECT NOTES STAT

[2]

ANDREWS DF, 1978, J ROY STAT SOC B MET, V40, P85

[3]

[Anonymous], 1988, NONLINEAR REGRESSION

[4]

BATES DM, 1980, J ROY STAT SOC B MET, V42, P1

[5] PARAMETER TRANSFORMATIONS FOR IMPROVED APPROXIMATE CONFIDENCE-REGIONS IN NON-LINEAR LEAST-SQUARES [J].

BATES, DM ;

WATTS, DG .

ANNALS OF STATISTICS, 1981, 9 (06) :1152-1167

[6]

BEALE EML, 1960, J ROY STAT SOC B, V22, P41

[7]

Belsley D. A., 2005, REGRESSION DIAGNOSTI

[8]

Box G., 1956, P I ELEC ENG B S1, V103, P100

[9]

Box G.E.P., 1992, BAYESIAN INFERENCE S

[10]

CLARKE GPY, 1980, J R STAT SOC B, V42, P227

← 1 2 3 →