Lopa, Samia
(2015)
Inference on quantile residual life for length-biased survival data.
Doctoral Dissertation, University of Pittsburgh.
(Unpublished)
Abstract
Length biased data occurs when a prevalent sampling is used to recruit subjects into a study that investigates the time from an initial event to a terminal event. Such data are usually left-truncated and right-censored. While there have been accurate and efficient methods to estimate the survival function, not much work has been done regarding the estimation of the residual life time distribution or the summary parameters such as the median and quantiles of the residual life. This dissertation proposes to make two new contributions. In the first part of the dissertation, we propose two ways to estimate the quantiles of the residual life time at fixed time points accounting for the length biased and censored nature of the data. We provide the asymptotic properties of these estimators and investigate them through simulation studies. Considering that the variances of these estimators require density estimation, we suggest an alternate approach taken by Jeong and others to obtain the confidence interval for the available residual function. We apply these methods to a breast cancer dataset from the National Surgical Adjuvant Breast and Bowel Project (NSABP).
n the second part of the dissertation, we propose a method for testing the equality of quantile residual life times from two different populations under prevalence sampling. This test can also be inverted to construct confidence intervals for the ratio or difference of two quantile residual life between two populations. We compare the performance of two methods, namely, the TPL and Huang and Qin methods via simulation. The results show that the proposed tests maintain Type I error. The test based on Huang and Qin survival estimator is more powerful than that of based on the TPL estimator. We apply our methods to test the equality of median residual life of breast cancer patients having recurrence and undergoing two different treatments.
Public health significance of this research is enormous. For a population experiencing certain disease such as cancer, it is important to estimate the quantiles of the residual life time at specific time points to assess the impact of a disease and an intervention strategy on the population. This dissertation will provide accurate and efficient methods for estimating these quantiles.
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| ETD Committee: |
|
| Date: |
28 January 2015 |
| Date Type: |
Publication |
| Defense Date: |
2 December 2014 |
| Approval Date: |
28 January 2015 |
| Submission Date: |
23 November 2014 |
| Access Restriction: |
2 year -- Restrict access to University of Pittsburgh for a period of 2 years. |
| Number of Pages: |
68 |
| Institution: |
University of Pittsburgh |
| Schools and Programs: |
School of Public Health > Biostatistics |
| Degree: |
PhD - Doctor of Philosophy |
| Thesis Type: |
Doctoral Dissertation |
| Refereed: |
Yes |
| Uncontrolled Keywords: |
Length-biased |
| Date Deposited: |
28 Jan 2015 16:45 |
| Last Modified: |
01 Jan 2017 06:15 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/23453 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[img]](https://d-scholarship.pitt.edu/style/images/fileicons/application_pdf.png)
![[feed]](/images/feed-icon-32x32.png){kind=icon}