|
Activity Number:
|
419
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 9, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Nonparametric Statistics
|
| Abstract - #305515 |
|
Title:
|
Inference on the Quantile Function under Left Truncation and Right Censoring
|
|
Author(s):
|
Sana Buhamra*+ and Noriah Al-Kandari
|
|
Companies:
|
Kuwait University and Kuwait University
|
|
Address:
|
P.O. Box 5969, Safat, Kuwait, 13060, Kuwait
|
|
Keywords:
|
quantile function ; shrinkage ; kernel estimation ; truncated and censored data ; Monte Carlo
|
|
Abstract:
|
We consider estimating the quantile function when data are left-truncated and right-censored (LTRC). Asymptotic statistical testing and estimation procedures are established for the quantile function from one sample and multiple-sample LTRC data. The proposed estimators are based on shrinkage estimation techniques assuming uncertain prior nonsample information on the value of the quantile. The asymptotic bias and risk of the proposed estimators are derived and compared with the benchmark estimator analytically and computationally. The proposed estimation strategy, which combines the sample and parameter information, performs better than a strategy based on sample information only. An application of the proposed methodology to the well-known Channing house data illustrates the theory.
|
