The theory behind sampling is to acquire knowledge of a parameter with the help of only partial information drawn from sample observations. A sample is expected to provide a good estimator of the unknown parameter or population. Given a sampling design, one of the fundamental issues in choosing a sample is the knowledge of the first and second order inclusion probabilities. This article introduces a new parametric family of sampling designs called determinantal sampling that can address all fundamental issues. The proposed sampling algorithms with given first and second order inclusion probabilities are described and statistical properties of the estimators based on the proposed sampling are studied. The proposed sampling design is applied to a real data set and results discussed.
展开▼
