Diagnostics high leverage cases and influential observations in regression analysis

被引：0

作者：

Zhao, JW ^{[1
]}

机构：

[1] Dongbei Univ Finance & Econ, Dept Stat, Dalian 116025, Peoples R China

来源：

8TH WORLD MULTI-CONFERENCE ON SYSTEMICS, CYBERNETICS, AND INFORMATICS, VOL XVI, PROCEEDINGS | 2004年

关键词：

case diagnostic; high leverage case; influential observation; wishart distribution; T-2; distribution; Wilks distribution;

D O I：

暂无

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

In this article, we first give the probability density function of Andrews-Pregibon's diagnostic statistic. Then we suggest two new diagnostic methods for high leverage cases and influential observations in regression analysis that they based on the multivariate standard distributions, such as, Wishart distribution, T-2 distribution, Wilks distribution, and so on. We find that these methods are very successful in assessing influence. They can diagnostic the high leverage cases and influential observations in regression analysis more easily.

引用

页码：417 / 422

页数：6

共 50 条

[1] Identification and classification of multiple outliers, high leverage points and influential observations in linear regression
Nurunnabi, A. A. M.
Nasser, M.
Imon, A. H. M. R.
[J]. JOURNAL OF APPLIED STATISTICS, 2016, 43 (03) : 509 - 525
[2] REGRESSION DIAGNOSTICS - AN EXPOSITORY TREATMENT OF OUTLIERS AND INFLUENTIAL CASES
BOLLEN, KA
JACKMAN, RW
[J]. SOCIOLOGICAL METHODS & RESEARCH, 1985, 13 (04) : 510 - 542
[3] Diagnostics of multiple group influential observations for logistic regression models
Coskun, Burcin
Alpu, O.
[J]. JOURNAL OF STATISTICAL COMPUTATION AND SIMULATION, 2019, 89 (16) : 3118 - 3136
[4] OUTLIERS, LEVERAGE OBSERVATIONS, AND INFLUENTIAL CASES IN FACTOR ANALYSIS: USING ROBUST PROCEDURES TO MINIMIZE THEIR EFFECT
Yuan, Ke-Hai
Zhong, Xiaoling
[J]. SOCIOLOGICAL METHODOLOGY, VOL 38, 2008, 38 : 329 - 368
[5] Leverage, residual, and interaction diagnostics for subsets of cases in least squares regression
Barrett, BE
Gray, JB
[J]. COMPUTATIONAL STATISTICS & DATA ANALYSIS, 1997, 26 (01) : 39 - 52
[6] Detection of Influential Observations in Spatial Regression Model Based on Outliers and Bad Leverage Classification
Baba, Ali Mohammed
Midi, Habshah
Adam, Mohd Bakri
Rahman, Nur Haizum Abd
[J]. SYMMETRY-BASEL, 2021, 13 (11):
[7] Leverage and influential observations on the Liu type estimator in the linear regression model with the severe collinearity
Eledum, Hussein
[J]. HELIYON, 2021, 7 (08)
[8] Diagnostic plot for the identification of high leverage collinearity-influential observations
Bagheri, Arezoo
Midi, Habshah
[J]. SORT, 2015, 39 (01): : 51 - 70
[9] Diagnostic plot for the identification of high leverage collinearity-influential observations
Bagheri, Arezoo
Midi, Habshah
[J]. SORT-STATISTICS AND OPERATIONS RESEARCH TRANSACTIONS, 2015, 39 (01) : 51 - 69
[10] Detecting influential observations in sliced inverse regression analysis
Prendergast, Luke A.
[J]. AUSTRALIAN & NEW ZEALAND JOURNAL OF STATISTICS, 2006, 48 (03) : 285 - 304

← 1 2 3 4 5 →