Using balanced sampling in creel surveys
Section 2. Balanced sampling
Suppose that is a finite population of size that is sampled with a design having selection probabilities given by If is an auxiliary variable known for all population units, then the sample is balanced on if the Horvitz-Thompson estimator for the total of is equal to the known total of In other words, for any balanced sample the following equation has to be satisfied,
For the surveys considered here, we balance on indicator variables equal to 1 if unit is of type and 0 otherwise. If all the units for which is equal to 1 have the same selection probability then equation (2.1) reduces to In this context the balancing equation simply requests that the number of sampled units of type is equal to its expectation,
To implement balanced sampling we use the cube method of Deville and Tillé (2004), and the extension of Hasler and Tillé (2014) to cope with highly stratified populations. In Section 4 this method is compared with the implementation of the rejection method proposed by Fuller (2009). In the context of this study, we are balancing on types of units; we want the sampled numbers of units for the types, to be equal to their expectations, under the sampling design. Under rejective sampling, the sample is said to be balanced if
where represents the design based covariance matrix of and is a tolerance value that determines the balancing condition. Samples that do not meet the balancing equation are simply rejected.
- Date modified: