Many estimators have been proposed for sampling designs with probabilities proportional to size and without replacement (PPSWOR sampling). Due to the lack of completeness of any non trivial sufficient statistic, no unique optimal estimator exists. Many of the existing "good" estimators are quite cumbersome to compute. In this paper a new estimator, based on a regression approach, is presented, together with a simple, easy to implement, recursive scheme of computation.
PPSWOR Sampling from Finite Populations: a Regression Approach to Estimation
ANDREATTA, GIOVANNI
1984
Abstract
Many estimators have been proposed for sampling designs with probabilities proportional to size and without replacement (PPSWOR sampling). Due to the lack of completeness of any non trivial sufficient statistic, no unique optimal estimator exists. Many of the existing "good" estimators are quite cumbersome to compute. In this paper a new estimator, based on a regression approach, is presented, together with a simple, easy to implement, recursive scheme of computation.
File in questo prodotto:
Non ci sono file associati a questo prodotto.
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
