A note on local influence based on normal curvature

被引:37
作者
Fung, WK [1 ]
Kwan, CW [1 ]
机构
[1] UNIV HONG KONG,DEPT STAT,HONG KONG,HONG KONG
来源
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-METHODOLOGICAL | 1997年 / 59卷 / 04期
关键词
local influence; normal curvature; scale invariance;
D O I
10.1111/1467-9868.00100
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
Object functions other than the likelihood displacement, such as a parameter estimate or a test statistic, can also be used in local influence analysis. The normal curvatures of these object functions have been studied in situations where the slopes were non-zero. In these situations, we show that the normal curvature is not scale invariant and thus ambiguous conclusions will be drawn. Comments on the application of the general normal curvature formula are presented.
引用
收藏
页码:839 / 843
页数:5
相关论文
共 10 条
[1]  
[Anonymous], 1982, Residuals and influence in regression
[2]   LOCAL INFLUENCE - A NEW APPROACH [J].
BILLOR, N ;
LOYNES, RM .
COMMUNICATIONS IN STATISTICS-THEORY AND METHODS, 1993, 22 (06) :1595-1611
[3]  
COOK RD, 1986, J ROY STAT SOC B MET, V48, P133
[4]   REGRESSION TRANSFORMATION DIAGNOSTICS USING LOCAL INFLUENCE [J].
LAWRANCE, AJ .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1988, 83 (404) :1067-1072
[5]   Sensitivity of Pearson's goodness-of-fit statistic in generalized linear models [J].
Lee, AH ;
Zhao, YJ .
COMMUNICATIONS IN STATISTICS-THEORY AND METHODS, 1996, 25 (01) :143-157
[6]  
Rousseeuw P. J., 2003, ROBUST REGRESSION OU
[7]   ASSESSING INFLUENCE ON PREDICTIONS FROM GENERALIZED LINEAR-MODELS [J].
THOMAS, W ;
COOK, RD .
TECHNOMETRICS, 1990, 32 (01) :59-65
[8]   TRANSFORMATION-MODEL DIAGNOSTICS [J].
TSAI, CL ;
WU, XZ .
TECHNOMETRICS, 1992, 34 (02) :197-202
[9]  
WU XZ, 1993, J ROY STAT SOC B MET, V55, P929
[10]   RESIDUAL SUM OF SQUARES AND MULTIPLE POTENTIAL, DIAGNOSTICS BY A 2ND-ORDER LOCAL APPROACH [J].
WU, XZ ;
LUO, Z .
STATISTICS & PROBABILITY LETTERS, 1993, 16 (04) :289-296