|
An Application of a Balanced Sampling Approach for Multi-Way Stratification Designs for Small Area Estimation
|
| Fateme Mashayekhi *, Hamidreza Navvabpour |
|
|
|
Abstract: (2507 Views) |
| When the purpose of sampling is to compute estimators for specific domains of population, it is possible to increase accuracy of domain estimators through stratifying population according to domains of interest and determining sample size of the domains. In this article stratified sampling design and balanced sampling design are introduced in which accuracy of domains estimators have been increased in design stage through considering domains as the strata. Also in estimating stage, utilizing modified estimators like synthetic and modified generalized regression estimator lead to more accurate estimates compare with Horvitz-Thompson estimator. Finally in a simulation study using data from 1390 establishment survey with 50 or more labor data, an application of balanced sampling approach for multi-way stratification designs for small area estimation has been explained where in synthetic estimator and modified generalized regression estimator with Horvitz-Thompson estimator for specific domains of the population are compared. |
|
| Keywords: Modified generalized regression estimator, Horvitz-Thompson estimator, synthetic estimator, balanced sampling, stratified sampling. |
|
|
Full-Text [PDF 373 kb]
(669 Downloads)
|
Type of Study: Applicable |
Subject:
Special
Received: 2017/01/28 | Accepted: 2018/07/7 | Published: 2018/09/25
|
|
|
|
|
|
