Heterogeneity models in lifetime data analysis

Abstract

Differences between experimental units, which cannot be explained by the available covariates, are frequently observed in studies based on time-to-event data. Failure to take into account these differences may have serious consequences that cannot be ignored. Consequently, developing methods for incorporating this heterogeneity into statistical models is a necessary objective of survival and reliability analysis, the sector of statistics that deals with the analysis of lifetime data. The basic tools for introducing heterogeneity are frailty and mixture models. The properties of frailty models, and in particular proportional hazards frailty models, are the first topic studied in this thesis. Diagnostic tests are developed for the correct specification of the model. Methods of estimation of the proportional hazards model and of mixtures are developed for the case where data have been collected with biased sampling designs. Finally, the proportional odds frailty model is developed and stu ...
show more

All items in National Archive of Phd theses are protected by copyright.

DOI
10.12681/eadd/16891
Handle URL
http://hdl.handle.net/10442/hedi/16891
ND
16891
Alternative title
Μοντέλα ετερογένειας στην ανάλυση δεδομένων διάρκειας ζωής
Author
Economou, Polychronis (Father's name: Dimitrios)
Date
2007
Degree Grantor
National Technical University of Athens (NTUA)
Committee members
Καρώνη-Ρίτσαρντσον Χρυσηίς
Κοκολάκης Γεώργιος
Κοκκουβίνος Χρήστος
Κούτρας Μάρκος
Μουστάκη Ειρήνη
Σπηλιώτης Ιωάννης
Χαραλαμπίδης Χαράλαμπος
Discipline
Natural Sciences
Mathematics
Keywords
Heterogeneity models; Shared frailty; Non shared frailty; Diagnostic tests; Biased sampling; Proportional ODDS frailty models
Country
Greece
Language
Greek
Description
226 σ., im.
Usage statistics
VIEWS
Concern the unique Ph.D. Thesis' views for the period 07/2018 - 07/2023.
Source: Google Analytics.
ONLINE READER
Concern the online reader's opening for the period 07/2018 - 07/2023.
Source: Google Analytics.
DOWNLOADS
Concern all downloads of this Ph.D. Thesis' digital file.
Source: National Archive of Ph.D. Theses.
USERS
Concern all registered users of National Archive of Ph.D. Theses who have interacted with this Ph.D. Thesis. Mostly, it concerns downloads.
Source: National Archive of Ph.D. Theses.
Related items (based on users' visits)