Journal of Official Statistics, Vol.21, No.4, 2005. pp. 543–570
Evaluation of Variance Approximations and Estimators in Maximum Entropy Sampling with Unequal Probability and Fixed Sample Size
Alina Matei and Yves Till
Abstract:Recent developments in survey sampling allow one to quickly draw samples with unequal probability, maximum entropy and fixed sample size. The joint inclusion probabilities can be computed exactly. For this sampling design, 7 approximations and 20 estimators of variance have been computed. A large set of simulations shows that knowledge of the joint inclusion probabilities is not necessary in order to obtain an accurate variance estimator.
Keywords:Variance, unequal probabilities, maximum entropy, fixed sample size, simulations
Copyright © Statistics Sweden 1996-2018. Open AccessISSN 0282-423XCreated and Maintained by OKS Group