Institutional Repository of Vilnius University
Multiple outliers identification method in accelerated failure time (aft) - regression models /
Abstract
dc:descriptionThe purpose of the study is to give modifications of outliers identification methods for normal linear regression to more general case of accelerated failure time (AFT) regression and compare them. In this work three outlier search methods are investigated: BP and David Gather (DG) based on one of two parameter estimation methods: ordinary least squares and robust. The objective of the first part is to modify BP outlier search method for accelerated failure time (AFT) regression models. The objective of the second part is to compare all three methods using data generation and to provide practical examples. In many situations, BP outlier search method identifies outliers better then both DG methods (ordinary least squares and robust). Analysis of real data examples confirms simulation results. Obtained results proved, that BP outlier search method can be useful for outliers search in accelerated failure time (AFT) regression model.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2021
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Sinkevičius, Deivydas,
- Contributors dc:contributor
-
- Bagdonavičius, Vilijandas
Subjects
dc:subject × 1Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- eng
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD81590249&prefLang=en_US