Journal article
An Exploratory Procedure for Identifying Influential Subgroups in Linear Regression
Quality & quantity, v 20(2-3), pp 235-248
01 Jan 1986
Featured in Collection : UN Sustainable Development Goals @ Drexel
Abstract
In some linear regression problems, samples may be unwittingly drawn from a population that is heterogeneous in respect to intercept. For such problems, an exploratory method is developed to identify influential subgroups of a heterogeneous population. Discriminant analysis is used to characterize, interpret, & validate subgroups. Four criteria are proposed to select a set of influential subgroups, which are then used in estimating a linear regression model. The method is illustrated through analysis of real data on political violence. 3 Tables, 2 Figures, 33 References. HA
Metrics
10 Record Views
Details
- Title
- An Exploratory Procedure for Identifying Influential Subgroups in Linear Regression
- Creators
- Lalit Aggarwal - Drexel University
- Publication Details
- Quality & quantity, v 20(2-3), pp 235-248
- Publisher
- Springer Nature
- Resource Type
- Journal article
- Language
- English
- Academic Unit
- [Retired Faculty]
- Web of Science ID
- WOS:A1986F168300008
- Scopus ID
- 2-s2.0-34247813332
- Other Identifier
- 991019173854704721
UN Sustainable Development Goals (SDGs)
This publication has contributed to the advancement of the following goals:
InCites Highlights
Data related to this publication, from InCites Benchmarking & Analytics tool:
- Web of Science research areas
- Social Sciences, Interdisciplinary
- Statistics & Probability