Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

Inference on conditional quantile residual life for censored survival data

Wu, Wen-Chi (2014) Inference on conditional quantile residual life for censored survival data. Doctoral Dissertation, University of Pittsburgh. (Unpublished)

[img]
Preview
PDF
Submitted Version

Download (622kB) | Preview

Abstract

For randomly censored data, the residual life function at a given time determines a life distribution of a subject survived up to that time point. In the situation where the data are censored, or where the underlying distribution is skewed, the quantile residual life function is preferred. A number of studies regarding the quantile residual lifetime have been conducted in the univariate settings by many professionals. However, when a pair of units are observed, i.e. a study of twins, or when patients experience two types of events, i.e. time to morbidity and time to mortality, a bivariate modelling of quantile residual lifetime subject to right censoring might be of utmost interest. In this dissertation, we develop the estimation of conditional quantile residual lifetime on semi-competing risks data. The proposed estimator is conditioning on the occurrence of the nonterminal event beyond time t. The covariate effects on specifc pairs of failure times are evaluated based on a log-linear regression on conditional
quantile residual lifetime for semi-competing risks data. Numerical studies demonstrate a reasonable performance of the estimator for moderate sample sizes. The proposed method
is applied to a study of breast cancer data from a phase III clinical trial.
Public Health Significance: In many survival studies, bivariate correlated failure times can be observed in a pair or in the same individual experiencing multiple failure times. It is of interest to know the additional time to failure of a surviving unit, when another unit is known to have failed at an earlier time. In this dissertation, the proposed estimator of the residual lifetime given the occurrence of a failure demonstrates the importance of lifetime expectancy that patients and their family seek to know before an onset of a new treatment.


Share

Citation/Export:
Social Networking:
Share |

Details

Item Type: University of Pittsburgh ETD
Status: Unpublished
Creators/Authors:
CreatorsEmailPitt UsernameORCID
Wu, Wen-Chiwenchi.wu82@gmail.com
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Thesis AdvisorJeong, Jong-Hyeonjeong@nsabp.pitt.eduJJEONG
Committee MemberWahed, AbduswahedA@edc.pitt.eduWAHED
Committee MemberLi, Ruosharul12@pitt.eduRUL12
Committee MemberCheng, Yuyucheng@pitt.eduYUCHENG
Date: 29 September 2014
Date Type: Publication
Defense Date: 18 July 2014
Approval Date: 29 September 2014
Submission Date: 18 July 2014
Access Restriction: 2 year -- Restrict access to University of Pittsburgh for a period of 2 years.
Number of Pages: 72
Institution: University of Pittsburgh
Schools and Programs: Graduate School of Public Health > Biostatistics
Degree: PhD - Doctor of Philosophy
Thesis Type: Doctoral Dissertation
Refereed: Yes
Uncontrolled Keywords: Residual life, Bivariate survival function, Kernel density estimator, Kaplan-Meier estimator, Semi-competing risks.
Related URLs:
Date Deposited: 29 Sep 2014 21:17
Last Modified: 15 Nov 2016 14:22
URI: http://d-scholarship.pitt.edu/id/eprint/22395

Metrics

Monthly Views for the past 3 years

Plum Analytics


Actions (login required)

View Item View Item