Voici les éléments 1 - 9 sur 9
Pas de vignette d'image disponible
Publication
Métadonnées seulement

Sampling Procedures for Coordinating Stratified Samples: Methods Based on Microstrata

2008-3-23, Nedyalkova, Desislava, Tillé, Yves, Pea, Johan

The aim of sampling coordination is to maximize or minimize the overlap between several samples drawn successively in a population that changes over time. Therefore, the selection of a new sample will depend on the samples previously drawn. In order to obtain a larger (or smaller) overlap of the samples than the one obtained by independent selection of samples, a dependence between the samples must be introduced. This dependence will emphasize (or limit) the number of common units in the selected samples. Several methods for coordinating stratified samples, such as the Kish & Scott method, the Cotton & Hesse method, and the Rivière method, have already been developed. Using simulations, we compare the optimality of these methods and their quality of coordination. We present six new methods based on permanent random numbers (PRNs) and microstrata. These new methods have the advantage of allowing us to choose between positive or negative coordination with each of the previous samples. Simulations are run to test the validity of each of them.

Pas de vignette d'image disponible
Publication
Métadonnées seulement

Sondage systématique et sondages à support minimal

2007-11, Qualité, Lionel, Tillé, Yves, Pea, Johan

Pas de vignette d'image disponible
Publication
Métadonnées seulement

A Review of Some Current Methods of Coordination of Stratified Samples. Introduction and Comparison of New Methods Based on Microstrata

2006-3-14, Nedyalkova, Desislava, Pea, Johan, Tillé, Yves

Pas de vignette d'image disponible
Publication
Métadonnées seulement

Sondage systématique et sondages à support minimal

2008, Pea, Johan, Qualité, Lionel, Tillé, Yves, Guilbert, Philippe, Haziza, David, Ruiz-Gazen, Anne, Tillé, Yves

Pas de vignette d'image disponible
Publication
Métadonnées seulement

Systematic sampling is a minimal support design

2007-3-23, Pea, Johan, Qualité, Lionel, Tillé, Yves

In order to select a sample in a finite population of N units with given inclusion probabilities, it is possible to define asamplingdesign on at most N samples that have a positive probability of being selected. Designs defined on minimal sets of samples are called minimum supportdesigns. It is shown that, for any vector of inclusion probabilities, systematicsampling always provides a minimum supportdesign. This property makes it possible to extensively compute the samplingdesign and the joint inclusion probabilities. Random systematicsampling can be viewed as the random choice of a minimum supportdesign. However, even if the population is randomly sorted, a simple example shows that some joint inclusion probabilities can be equal to zero. Another way of randomly selecting a minimum supportdesign is proposed, in such a way that all the samples have a positive probability of being selected, and all the joint inclusion probabilities are positive.

Pas de vignette d'image disponible
Publication
Métadonnées seulement

Evaluation of Coordination Methods for Simple and Stratified Samples

2006, Pea, Johan, Tillé, Yves

Pas de vignette d'image disponible
Publication
Métadonnées seulement

Sampling procedures for coordinating stratified samples: methods based on microstrata

2007-12, Nedyalkova, Desislava, Tillé, Yves, Pea, Johan

Pas de vignette d'image disponible
Publication
Métadonnées seulement

Evaluation of Coordination Methods for Simple and Stratified Samples

2006-4, Tillé, Yves, Pea, Johan

Vignette d'image
Publication
Accès libre

Sampling Procedures for Coordinating Stratified Samples : Methods Based on Microstrata

, Nedyalkova, Desislava, Tillé, Yves, Pea, Johan

The aim of sampling coordination is to maximize or minimize the overlap between several samples drawn successively in a population that changes over time. Therefore, the selection of a new sample will depend on the samples previously drawn. In order to obtain a larger (or smaller) overlap of the samples than the one obtained by independent selection of samples, a dependence between the samplesmust be introduced. This dependence will emphasize (or limit) the number of common units in the selected samples. Several methods for coordinating stratified samples, such as the Kish & Scott method, the Cotton & Hesse method, and the Rivi`ere method, have already been developed. Using simulations, we compare the optimality of these methods and their quality of coordination. We present six new methods based on permanent random numbers (PRNs) and microstrata. These new methods have the advantage of allowing us to choose between positive or negative coordination with each of the previous samples. Simulations are run to test the validity of each of them.