• Home
  1. Home
  2. Criterion Mixing Lack of Fit and Multicolinearity for Variable Selection in Regression
Report
English

Criterion Mixing Lack of Fit and Multicolinearity for Variable Selection in Regression

Contributors

Rolle, Jean-Daniel

Collection

  • Cahiers de recherche; 1999.11

Publication date

1999

Abstract

We describe in this paper a method allowing to order submodels in linear regression. A real function is attached to each submodel, allowing to graphically compare and order them. Our procedure defines an objective function depending on two factors (lack of fit and multicolinearity) with the property that the minimum of this function over subsets of the regressors determines the "best" such subset of variables. The latter will be found by systematically examining all possible subsets having at least two regressors. The method is built in such a way that it allows a graphical inspection of the submodels. Although it is presented here only in the framework of linear regression with a single, continuous, response variable, the procedure may be extended in a natural way to more general regression models. Simulation and implementation on various data sets gave very satisfying results

UNIGE affiliation entities

Citation (ISO format)

ROLLE, Jean-Daniel. Criterion Mixing Lack of Fit and Multicolinearity for Variable Selection in Regression. 1999

Identifiers

  • PID : unige:5873
526views
0downloads

Technical informations

Creation

15/04/2010 12:20:54

First validation

15/04/2010 12:20:54

Update

14/03/2023 15:26:56

Status update

14/03/2023 15:26:56

Last indexation

29/10/2024 14:25:12
All rights reserved by Archive ouverte UNIGE and the University of Geneva
OpenAIRE Indexed & Validated