|
Activity Number:
|
166
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Nonparametric Statistics
|
| Abstract - #304047 |
|
Title:
|
Wavelet Function Estimation for Right-Censored Data
|
|
Author(s):
|
Jeong-Ran Lee*+ and Hee-Seok Oh
|
|
Companies:
|
Seoul National University and Seoul National University
|
|
Address:
|
599, Gwanak-ro, Gwanak-gu, Seoul, International, 151-742, South Korea
|
|
Keywords:
|
Binning process ; Block thresholding ; Density estimation ; Right censored data ; Root-unroot transform ; Wavelets
|
|
Abstract:
|
Density estimation and nonparametric regression are two fundamental nonparametric problems. There are some useful tricks for converting a density estimation problem into a regression problem. Then we can use all the regression methods which are convenient to implement and give an effective density estimator enjoys a high degree of adaptivity. The algorithm is particularly convenient for the use of wavelet regression methods because with no difficulty it can provide the binary number of equally spaced regression observations for which a wavelet method is most suited. We discuss wavelet methods for estimating subdensity and hazard rate function of right censored data. By making some modifications to the binning process and root-unroot algorithm, we propose a spatial adaptive method which performs well when the true function has some special local structures.
|
