Skip to Main content Skip to Navigation

Méthodes de Bootstrap en population finie

Abstract : This Phd deals with Bootstrap methods for finite population sampling. The first chapter introduces some bases about sampling and gives an overview of the main variance estimation techniques. A remind on Bootstrap methods for simple random sampling is given in chapter 2, and two new methods are introduced. A Bootstrap algorithm for unequal probability sampling is proposed in chapter 3, and shown to be consistent for variance estimation of plug-in statistics in case of large entropy sampling designs. Balanced sampling is presented in chapter 4, and a fast algorithm is proposed. Former Bootstrap algorithm is shown to be consistent as well in case of variance estimation for a maximum entropy balanced sampling design. Cases of complex sampling designs or reweighting strategies are discussed in chapter 5. An application to the French Renovated Census is given in chapter 6
Document type :
Complete list of metadata

Cited literature [199 references]  Display  Hide  Download
Contributor : Gaid Thirion Connect in order to contact the contributor
Submitted on : Friday, March 28, 2008 - 10:20:48 AM
Last modification on : Friday, August 5, 2022 - 2:49:41 PM
Long-term archiving on: : Thursday, May 20, 2010 - 8:04:41 PM


  • HAL Id : tel-00267689, version 1



Guillaume Chauvet. Méthodes de Bootstrap en population finie. Mathématiques [math]. ENSAI, 2007. Français. ⟨tel-00267689⟩



Record views


Files downloads