Conditional Distribution-Based Downweighting for Robust Estimation of Logistic Regression Models

Weichang Yu () and Howard D. Bondell ()
Additional contact information
Weichang Yu: University of Melbourne, School of Mathematics and Statistics
Howard D. Bondell: University of Melbourne, School of Mathematics and Statistics

A chapter in Robust and Multivariate Statistical Methods, 2023, pp 349-364 from Springer

Abstract: Abstract We propose a new approach to robust estimation for logistic regression based on the conditional covariate distribution given the binary response. Instead of downweighting observations based on the marginal distribution often adopted in common approaches (e.g., the Mallows class), we propose using the conditional distributions that align with a case-control perspective of binary regression. We justify our proposed weighting scheme by showing that our method based on this new perspective leads to sensible weights in scenarios where the existing methods perform poorly. Through simulated and real datasets, we show that our proposed estimator achieves superior performance over the existing estimators in terms of both robustness and efficiency.

Keywords: Case-control; Efficiency; M-estimation; Robustness (search for similar items in EconPapers)
Date: 2023
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-3-031-22687-8_16

Ordering information: This item can be ordered from
http://www.springer.com/9783031226878

DOI: 10.1007/978-3-031-22687-8_16

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().