|
Activity Number:
|
189
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, August 7, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Nonparametric Statistics
|
| Abstract - #305961 |
|
Title:
|
Ranked Set Sample Inference under the Constraint of Stochastic Ordering of Judgment Ranking Classes
|
|
Author(s):
|
Omer Ozturk*+
|
|
Companies:
|
The Ohio State University
|
|
Address:
|
1958 Neil Ave., Columbus, OH, 43210,
|
|
Keywords:
|
isotonic regression ; imperfect ranking ; perfect ranking ; consistent estimator ; ranking model
|
|
Abstract:
|
In ranked set sampling procedure, ranking error is almost an unavoidable reality. Thus, a reasonable statistical inference should address the validity of the procedures under imperfect ranking. In this talk, we will develop statistical inference for the estimation of the cumulative distribution function of the underlying distribution under the stochastic order constraint of the judgment class distributions. We will show that the CDF estimator of judgment classes are consistent for the CDF of the ranking class distributions. We next will use these estimators to develop statistical inference for ranked set samples that may contain ranking errors. Examples will be provided where these estimators can be used to draw inference for a characteristic of the underlying distribution.
|
