ON ADDRESSING WITH TIES IN SURVIVAL DATA USING FUZZY THEORY

Read full text Summarise with VIRAT
Authors: R Jaisankar, Haripriya. H
Cited by: 0
References: 10

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

The occurrence of one or more ties is quite common in any type of survival data that may arise when one or more subjects experience the event or censored at a same time point. The presence of ties may induce certain challenges both in analysis and interpretation of survival methods as many of them assume continuous event times. Many researchers have proposed certain procedures to handle ties, which consist of increasing the time resolution or treating tied observations as discrete or making adjustment in the computation of survival probabilities assuming tied events occur simultaneously, or by considering all possible orderings of events that occur at the same time and hence calculating the exact partial likelihood (Breslow, Efron). Some authors recommend random jittering by adding some noise to break the ties artificially. However, such methods do not guarantee the precision of survival estimates and sometimes may lead to even biased when the datasets are larger with many ties. This paper proposes a fuzzy oriented tie-handling method which administers the use of fuzzy numbers both for normal and tied observations. A comparative study has also been made on the application of the proposed method on Cox regression.

Fields of Study & Subjects

Fuzzy Systems and Optimization Statistical Distribution Estimation and Applications Probability and Statistical Research

Publication details

Journal / source
Reliability: Theory & Applications
Publication date
1 January 2026
Language
EN