Diagnostics in logistic regression models

被引：10

作者：

Sen Roy, Sugata ^{[1
]}

Guria, Sibnarayan ^{[2
]}

机构：

[1] Univ Calcutta, Dept Stat, Kolkata 700019, W Bengal, India

[2] Bidhannagar Coll, Dept Stat, Kolkata 700064, W Bengal, India

来源：

JOURNAL OF THE KOREAN STATISTICAL SOCIETY | 2008年 / 37卷 / 02期

关键词：

deletion of observation; deviance; logistic regression; maximum likelihood estimator;

D O I：

10.1016/j.jkss.2007.03.001

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 [统计学]; 070103 [概率论与数理统计]; 0714 [统计学];

摘要：

In this paper we study the diagnostics of a logistic regression model using the deletion of observation technique. The model is fitted using the maximum likelihood method and the changes in the estimates and the deviance are observed when the model is refitted after deleting an observation. Expressions are derived so that it is not necessary to re-run the regression after each deletion, thereby considerably saving the computational time. (C) 2008 The Korean Statistical Society. Published by Elsevier Ltd. All rights reserved.

引用

页码：89 / 94

页数：6

共 7 条

[1]

Belsley D. A., 1980, REGRESSION DIAGNOSTI

[2]

COLLETT D, 1998, ENCY BIOSTATISTICS, P350

[3]

Cook R. D., 1982, Residuals and influence in regression

[4]

COOK RD, 1986, J ROY STAT SOC B MET, V48, P133

[5]

LOGISTIC-REGRESSION DIAGNOSTICS [J].

PREGIBON, D .

ANNALS OF STATISTICS, 1981, 9 (04) :705-724

[6]

THOMAS W, 1989, BIOMETRIKA, V76, P741

[7]

GENERALIZED LINEAR-MODEL DIAGNOSTICS USING THE DEVIANCE AND SINGLE CASE DELETIONS [J].

WILLIAMS, DA .

APPLIED STATISTICS-JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES C, 1987, 36 (02) :181-191

← 1 →