Asymptotic normality for two-stage sampling from a finite population (Q1113228): Difference between revisions

We consider two-stage sampling from a finite population, and associated estimators of the population total, in a general setting which includes most two-stage procedures in the literature. The main result gives general conditions for asymptotic normality of the estimators. The proof is based on a martingale central limit theorem. It is indicated how the result can be extended to multi-stage procedures.