This paper deals with the estimation of the mean in soil surveys. After a brief recall of the two main approaches based on randomization distribution, on one side, and on the regionalised variable theory, on the other side, we propose an estimator assisted by a spline smoother of the population data under a desgin based perspective. The properties of the estimator are stated and some results from an ongoing simulation study carried out to investigate its performance in terms of relative bias and efficiency is quoted. The proposed estimator can represent a way of compromise between design-based and model-based paradigms taking advantages from both approaches.
Design-based and model-based inference in soil sciences: a way of reconciliation
CICCHITELLI, Giuseppe;MONTANARI, Giorgio Eduardo
2010
Abstract
This paper deals with the estimation of the mean in soil surveys. After a brief recall of the two main approaches based on randomization distribution, on one side, and on the regionalised variable theory, on the other side, we propose an estimator assisted by a spline smoother of the population data under a desgin based perspective. The properties of the estimator are stated and some results from an ongoing simulation study carried out to investigate its performance in terms of relative bias and efficiency is quoted. The proposed estimator can represent a way of compromise between design-based and model-based paradigms taking advantages from both approaches.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
