Journals Information
Mathematics and Statistics Vol. 13(5), pp. 413 - 419
DOI: 10.13189/ms.2025.130517
Reprint (PDF) (343Kb)
The Identification of Influential Groups in Linear Regression Models via an Influence Matrix Approach
Tobias Ejiofor Ugah 1, Anichebe Gregory Emeka 2, Ezeora Nnamdi Johnson 2, Caroline Ngozi Asogwa 2,*, Adaora Angela Obayi 2, Uchenna Charity Onwuamaeze 1, Emmanuel Ikechukwu Mba 1, Ifeoma Christy Mba 3, Egbo Mary Nkechinyere 4, Comfort Njideka Ekene-Okafor 5
1 Department of Statistics, Faculty of Physical Sciences, University of Nigeria, Nsukka, Nigeria
2 Department of Computer Science, Faculty of Physical Sciences, University of Nigeria, Nsukka, Nigeria
3 Department of Economics, Faculty of Social Sciences, University of Nigeria, Nsukka, Nigeria
4 Department of Statistics, Chukwuemeka Odumegwu Ojukwu University, Uli, Anambra, Nigeria
5 Department of Computer Science/Mathematics, Faculty of Natural Sciences and Environmental Studies, Godfrey Okoye University, Enugu, Nigeria
ABSTRACT
The Cook's distance measure is a prominent diagnostic tool for influence measure in linear regression diagnostics. Many authors have studied it, and the main focus is on its use for detection of a single influential observation in linear regression. In this work, we propose a standardized version of it and extend the single-case form to flag influential subsets. The proposed method uses diagonal and off-diagonal elements of a normalized influence matrix 
. The main diagonal elements of consist of the standardized univariate Cook statistics, and the off-diagonal elements consist of useful statistics that can detect influential subsets. A scattergram of the off-diagonal components of is drawn, and bounds (lower and upper bounds) are imposed on it. These bounds form the main artillery for detecting influential subsets. One of the glaring advantages of the approach is that it facilitates the identification of influential subsets that would be lost if only the main diagonal entries of are explored. The method is effective and computationally simple to apply, especially where more complex methods are not easy to implement, because they are computationally intensive. Analysis of well-known real-life data sets in linear regression diagnostics is used to illustrate the application and usefulness of the proposed method.
KEYWORDS
Critical Values, Bonferroni Inequality, Test Statistic, Studentized Residual, Hat Matrix, Leverage
Cite This Paper in IEEE or APA Citation Styles
(a). IEEE Format:
[1] Tobias Ejiofor Ugah , Anichebe Gregory Emeka , Ezeora Nnamdi Johnson , Caroline Ngozi Asogwa , Adaora Angela Obayi , Uchenna Charity Onwuamaeze , Emmanuel Ikechukwu Mba , Ifeoma Christy Mba , Egbo Mary Nkechinyere , Comfort Njideka Ekene-Okafor , "The Identification of Influential Groups in Linear Regression Models via an Influence Matrix Approach," Mathematics and Statistics, Vol. 13, No. 5, pp. 413 - 419, 2025. DOI: 10.13189/ms.2025.130517.
(b). APA Format:
Tobias Ejiofor Ugah , Anichebe Gregory Emeka , Ezeora Nnamdi Johnson , Caroline Ngozi Asogwa , Adaora Angela Obayi , Uchenna Charity Onwuamaeze , Emmanuel Ikechukwu Mba , Ifeoma Christy Mba , Egbo Mary Nkechinyere , Comfort Njideka Ekene-Okafor (2025). The Identification of Influential Groups in Linear Regression Models via an Influence Matrix Approach. Mathematics and Statistics, 13(5), 413 - 419. DOI: 10.13189/ms.2025.130517.