Activity Number:
|
309
|
Type:
|
Contributed
|
Date/Time:
|
Wednesday, August 14, 2002 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Survey Research Methods*
|
Abstract - #301507 |
Title:
|
The Horvitz-Thompson Predictor in Random Fields Sampling
|
Author(s):
|
Jose Rodriguez*+
|
Affiliation(s):
|
Universidad de Guanajuata
|
Address:
|
A[artado Postal 402, Guanajuata, GTO, International, 36000, MEXICO
|
Keywords:
|
Horvitz-Thompson estimator ; spatial sampling ; random fields sampling ; optimal sampling designs
|
Abstract:
|
The Horvitz-Thompson estimator is commonly used in finite population sampling. It is also used in spatial sampling when the characteristic of interest is considered as a deterministic spatial function. An extension of this estimator is developed for random fields sampling. The Horvitz-Thompson predictor is deduced as a special case from the nonparametric predictors of the integral of a random field. Optimal random sampling designs are obtained for simple and stratified sampling. The mean square error of each predictor is deduced, and then the unbiased estimator for this MSE is proposed.
|