In *Subjective and Objective Bayesian Statistics: Principles, Models, and Applications, Second Edition*, pages 336–358. John Wiley & Sons, Inc., Hoboken, NJ, USA, November, 2002.

doi abstract bibtex

doi abstract bibtex

[Excerpt: Introduction] Hierarchical modeling is a widely used approach to building complex models by specifying a series of more simple conditional distributions. It naturally lends itself to Bayesian inference, especially using modern tools for Bayesian computation. In this chapter we first present essential concepts of hierarchical modeling, and then suggest its generality by presenting a series of widely used specific models. [...] [\n] [...] [Summary] In this chapter we have introduced hierarchical modeling as a very general approach to specifying complex models through a sequence of more simple stages. Hierarchical models are useful for modeling coIlections of observations with a simple or complex exchangeability structure. Using fully Bayesian approaches, both general parameters (characterizing the entire population) and parameters specific to individual units (as in smaltarea estimation or profiling) can be estimated. Modem Bayesian computational approaches are well adapted to estimation of such models. [...]

@incollection{zaslavskyHierarchicalBayesianModeling2002, title = {Hierarchical {{Bayesian}} Modeling}, booktitle = {Subjective and {{Objective Bayesian Statistics}}: {{Principles}}, {{Models}}, and {{Applications}}, {{Second Edition}}}, author = {Zaslavsky, Alan M.}, editor = {Press, S. James}, year = {2002}, month = nov, pages = {336--358}, publisher = {{John Wiley \& Sons, Inc.}}, address = {{Hoboken, NJ, USA}}, doi = {10.1002/9780470317105.ch14}, abstract = {[Excerpt: Introduction] Hierarchical modeling is a widely used approach to building complex models by specifying a series of more simple conditional distributions. It naturally lends itself to Bayesian inference, especially using modern tools for Bayesian computation. In this chapter we first present essential concepts of hierarchical modeling, and then suggest its generality by presenting a series of widely used specific models. [...] [\textbackslash n] [...] [Summary] In this chapter we have introduced hierarchical modeling as a very general approach to specifying complex models through a sequence of more simple stages. Hierarchical models are useful for modeling coIlections of observations with a simple or complex exchangeability structure. Using fully Bayesian approaches, both general parameters (characterizing the entire population) and parameters specific to individual units (as in smaltarea estimation or profiling) can be estimated. Modem Bayesian computational approaches are well adapted to estimation of such models. [...]}, isbn = {978-0-470-31710-5}, keywords = {*imported-from-citeulike-INRMM,~INRMM-MiD:c-14546817,~to-add-doi-URL,bayesian,computational-science,mathematics,modelling,statistics}, lccn = {INRMM-MiD:c-14546817} }

Downloads: 0