A diagnostic method for simultaneous feature selection and outlier identification in linear regression

Authors:
Rajiv S. Menjoge;Roy E. Welsch
Affiliations:
Operations Research Center, M.I.T., Cambridge, 02139, MA, United States;Sloan School of Management, M.I.T., Cambridge, MA, United States
Venue:
Computational Statistics & Data Analysis
Year:
2010

Citing 4
Cited 1

Robust regression and outlier detection

Robust regression and outlier detection
A method for simultaneous variable selection and outlier identification in linear regression

Computational Statistics & Data Analysis
Ranking a random feature for variable and feature selection

The Journal of Machine Learning Research
Detecting multiple outliers in linear regression using a cluster method combined with graphical visualization

Computational Statistics

Editorial: Special issue on variable selection and robust procedures

Computational Statistics & Data Analysis

Quantified Score

Hi-index	0.03

Visualization

Abstract

A diagnostic method along the lines of forward search is proposed to simultaneously study the effect of individual observations and features on the inferences made in linear regression. The method operates by appending dummy variables to the data matrix and performing backward selection on the augmented matrix. It outputs sequences of feature-outlier combinations which can be evaluated by plots similar to those of forward search and includes the capacity to incorporate prior knowledge, in order to mitigate issues such as collinearity. It also allows for alternative ways to understand the selection of the final model. The method is evaluated on five data sets and yields promising results.