|
Activity Number:
|
420
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #304912 |
|
Title:
|
The Franklin's Randomized Response Model for Two Sensitive Attributes
|
|
Author(s):
|
Cheng C. Chen*+ and Sarjinder Singh
|
|
Companies:
|
Texas A&M University-Kingsville and Texas A&M University-Kingsville
|
|
Address:
|
Department of Mathematics, Kingsville, TX, 78363,
|
|
Keywords:
|
Franklin's randomized response model ; Estimation of proportion ; Sensitive attributes ; Higher order moments
|
|
Abstract:
|
In this paper, we suggest a new method to estimate the proportions of two non-overlapping sensitive attributes using the famous method of Franklin (1990:Communications in Statistics-Theory and Methods). The extension is based on the recent work of Singh and Chen (2009: Journal of the Statistical Planning and Inference) where they suggest to utilize higher order moments of the scrambling variable to estimate single proportion of a sensitive attribute in survey sampling. The most important point is that here we use information from one sample to estimate proportions of two sensitive attributes. An attempt has been made to extend the proposed estimators to Bayesian and Empirical Bayes estimation techniques. Simulation studies are also performed to see the performance of the proposed estimators with their competitors.
|
