Partenaires

CNRS
Logo tutelle
Logo tutelle
Logo tutelle


Rechercher

Sur ce site

Sur le Web du CNRS


Accueil du site > Séminaires > Probabilités Statistiques et réseaux de neurones > Sampling using Ranked Sets : some results in finite population inference

Vendredi 2 février 2007 à 11h00

Sampling using Ranked Sets : some results in finite population inference

Carlos Bouza (Université La Havane, Cuba), bouza@matcom.uh.cu

Résumé : Traditionally simple random sampling is considered as the token for selecting samples. During the last decade Ranked Set Sampling has been considered as an alternative to purely random selection. This design is based on the use of a two-stage model. Random samples are selected under the with replacement mechanism. The selected units are ordered (ranked). Each order statistic is observed once. This process can be repeated or not. The measurement of the random variable is made considering the order statistics of the samples.We present a review of the most significant results in this theme and some open problems related with this sampling design are quoted. We discuss in detail results in the use of ranked set sampling for estimating ratios and difference of means (Bouza, C. N. (2001b) : Model assisted ranked survey sampling. Biometrical J., 43, 249-259 and Bouza, C. N. (2001c) : Ranked set sampling for estimating the differences of means. Investigación Operacional, 22, 154-162.), for solving the problems present under missing observations (Bouza, C. N. (2001a) : Random set sampling with non-responses. Rev. Mat. Est. S. Paulo, 19, 297-308, Bouza , C. N. (2002a) : Estimation of the mean in ranked set sampling with non-responses. Metrika, 56, 171-179.) and the study of the gains in accuracy due to this design when we deal with randomized mechanisms for reducing the response bias in sensitive questions.

Dans la même rubrique :