BAYESIAN DESIGN OF SINGLE, DOUBLE, AND SEQUENTIAL STRATIFIED SAMPLING FOR ESTIMATING PROPORTION IN FINITE POPULATIONS.
Abstract
The problem of optimal allocation in stratified sampling to estimate a vector of proportions is studied from the Bayesian point of view. A linear function of the proportions is analyzed to make the problem more manageable. In particular, an investigation of double and sequential Bayes stratified sampling is performed. The methodology is outlined and two special cases are worked out: the cases of binomial and of uniform prior distributions of the number of elements in the strata having a certain attribute. The Bayesian efficiency of inverse sequential stratified sampling is studied in relation to the Bayesian stratified simple random sampling. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 22, 1967
- Accession Number
- AD0660045
Entities
People
- S. Zacks
Organizations
- Stanford University