|
Activity Number:
|
52
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Sunday, July 29, 2007 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Section on Nonparametric Statistics
|
| Abstract - #309389 |
|
Title:
|
Optimal Allocation in the Presence of Missing Data
|
|
Author(s):
|
Jessica Kohlschmidt*+ and Elizabeth A. Stasny and Douglas Wolfe
|
|
Companies:
|
The Ohio State University and The Ohio State University and The Ohio State University
|
|
Address:
|
404 Cockins Hall, Columbus, OH, 43210,
|
|
Keywords:
|
Ranked Set Sampling ; Missing data ; Optimal Allocation ; MCAR ; MAR
|
|
Abstract:
|
Missing data is a natural consequence of sampling. Some researchers ignore the missing data. If the data is missing completely at random, this procedure provides meaningful estimates. In many situations, the missingness in the data is due to a phenomenon occurring in the population. In ranked set sampling, we stratify the data by ranks. Many times the missingness in the population varies depending on which rank the observation came from. We will show the optimal allocation when the missing data parameters are known. Then we extend this to the case where the missing data parameters are unknown, a more convincing scenario. We explore the effect of the varying costs for collecting observations in each rank and its effect on the optimal allocation.
|
