Date of Award

1-1-2009

Document Type

Campus Access Thesis

Department

Epidemiology and Biostatistics

Sub-Department

Biostatistics

First Advisor

Matteo Bottai

Abstract

Laplace regression is a quantile regression method for analyzing censored data where the censored observations are assumed to come from an asymmetric Laplace distribution. In this simulation study the performance of Laplace regression was compared to Portnoy and Peng-Huang's methods, two other quantile regression methods for survival data. Performance of the methods was assessed primarily based on coverage percentage and interval length. In the simulation study Laplace regression outperformed both Portnoy and Peng-Huang's methods, showing larger coverage percentages with interval lengths that were generally comparable.

Share

COinS