|
Abstract:
|
Post-stratification, or sample balancing, or raking, is widely utilized in survey research to weighting a sample data for a better correspondence to Census or other known population quotas. Cross-tables of counts are mostly used in the Deming-Stephan iterative proportional fitting to find the weights for adjusting data to known margins. A bi-criteria objective for finding weights with minimum variance yields a solution with maximum effective sample size. This model can be expressed as a ridge regression, which is applied to the original data, without its collapsing to cross-tables. Linear and nonlinear parameterization models are studied. The explicit regression solution allows to study the weighting analytically, that helps to interpret and improve the sample balance results.
|
