The performance of classical and robust logistic regression estimators in the presence of outliers

It is now evident that the estimation of logistic regression parameters, using Maximum Likelihood Estimator(MLE), suffers a huge drawback in the presence of outliers. An alternative approach is to use robust logistic regression estimators, such as Mallows type leverage dependent weights estimator (...

Full description

Bibliographic Details
Main Authors: Midi, Habshah, Ariffin @ Mat Zin, Syaiba Balqish
Format: Article
Language:English
Published: Universiti Putra Malaysia Press 2012
Online Access:http://psasir.upm.edu.my/id/eprint/40467/1/16.%20The%20Performance%20of%20Classical%20and%20Robust%20Logistic%20Regression.pdf
_version_ 1825949606966460416
author Midi, Habshah
Ariffin @ Mat Zin, Syaiba Balqish
author_facet Midi, Habshah
Ariffin @ Mat Zin, Syaiba Balqish
author_sort Midi, Habshah
collection UPM
description It is now evident that the estimation of logistic regression parameters, using Maximum Likelihood Estimator(MLE), suffers a huge drawback in the presence of outliers. An alternative approach is to use robust logistic regression estimators, such as Mallows type leverage dependent weights estimator (MALLOWS, Conditionally Unbiased Bounded Influence Function estimator (CUBIF), Bianco and Yohai estimator (BY), and Weighted Bianco and Yohai estimator (WBY). This paper investigates the robustness of the preceding robust estimators by using real data sets and Monte Carlo simulations. The results indicate that the MLE behaves poorly in the presence of outliers. On the other hand, the WBY estimator is more efficient than the other existing robust estimators. Thus, it is suggested that the WBY estimator be employed when outliers are present in the data to obtain a reliable estimate.
first_indexed 2024-03-06T08:47:09Z
format Article
id upm.eprints-40467
institution Universiti Putra Malaysia
language English
last_indexed 2024-03-06T08:47:09Z
publishDate 2012
publisher Universiti Putra Malaysia Press
record_format dspace
spelling upm.eprints-404672015-11-04T04:01:37Z http://psasir.upm.edu.my/id/eprint/40467/ The performance of classical and robust logistic regression estimators in the presence of outliers Midi, Habshah Ariffin @ Mat Zin, Syaiba Balqish It is now evident that the estimation of logistic regression parameters, using Maximum Likelihood Estimator(MLE), suffers a huge drawback in the presence of outliers. An alternative approach is to use robust logistic regression estimators, such as Mallows type leverage dependent weights estimator (MALLOWS, Conditionally Unbiased Bounded Influence Function estimator (CUBIF), Bianco and Yohai estimator (BY), and Weighted Bianco and Yohai estimator (WBY). This paper investigates the robustness of the preceding robust estimators by using real data sets and Monte Carlo simulations. The results indicate that the MLE behaves poorly in the presence of outliers. On the other hand, the WBY estimator is more efficient than the other existing robust estimators. Thus, it is suggested that the WBY estimator be employed when outliers are present in the data to obtain a reliable estimate. Universiti Putra Malaysia Press 2012 Article PeerReviewed application/pdf en http://psasir.upm.edu.my/id/eprint/40467/1/16.%20The%20Performance%20of%20Classical%20and%20Robust%20Logistic%20Regression.pdf Midi, Habshah and Ariffin @ Mat Zin, Syaiba Balqish (2012) The performance of classical and robust logistic regression estimators in the presence of outliers. Pertanika Journal of Science & Technology, 20 (2). pp. 313-325. ISSN 0128-7680; ESSN: 2231-8526 http://www.pertanika.upm.edu.my/Pertanika%20PAPERS/JST%20Vol.%2020%20%282%29%20Jul.%202012/09%20Pg%20313-325.pdf
spellingShingle Midi, Habshah
Ariffin @ Mat Zin, Syaiba Balqish
The performance of classical and robust logistic regression estimators in the presence of outliers
title The performance of classical and robust logistic regression estimators in the presence of outliers
title_full The performance of classical and robust logistic regression estimators in the presence of outliers
title_fullStr The performance of classical and robust logistic regression estimators in the presence of outliers
title_full_unstemmed The performance of classical and robust logistic regression estimators in the presence of outliers
title_short The performance of classical and robust logistic regression estimators in the presence of outliers
title_sort performance of classical and robust logistic regression estimators in the presence of outliers
url http://psasir.upm.edu.my/id/eprint/40467/1/16.%20The%20Performance%20of%20Classical%20and%20Robust%20Logistic%20Regression.pdf
work_keys_str_mv AT midihabshah theperformanceofclassicalandrobustlogisticregressionestimatorsinthepresenceofoutliers
AT ariffinmatzinsyaibabalqish theperformanceofclassicalandrobustlogisticregressionestimatorsinthepresenceofoutliers
AT midihabshah performanceofclassicalandrobustlogisticregressionestimatorsinthepresenceofoutliers
AT ariffinmatzinsyaibabalqish performanceofclassicalandrobustlogisticregressionestimatorsinthepresenceofoutliers