Skip to Main content Skip to Navigation
Book sections

Bayesian analysis of structural equation models using parameter expansion

Abstract : Structural Equation Models with latent variables (SEM) are hypothetical constructs used to represent causality relationships in data, where the observed correlation structure is transferred into the correlation structure of latent variables. In this paper a Bayesian analysis of SEM is proposed using parameter expansion to overcome identi fiability issues. An original use of posterior draws from latent variables is proposed to model expert knowledge in uncertainty analysis.
Document type :
Book sections
Complete list of metadata

Cited literature [11 references]  Display  Hide  Download
Contributor : Laboratoire Cedric <>
Submitted on : Thursday, April 2, 2020 - 2:21:36 PM
Last modification on : Friday, February 5, 2021 - 4:16:34 PM


Files produced by the author(s)


  • HAL Id : hal-01125864, version 1
  • PRODINRA : 313055


Séverine Demeyer, Jean-Louis Foulley, Nicolas Fischer, Gilbert Saporta. Bayesian analysis of structural equation models using parameter expansion. Mireille Gettler Summa; Leon Bottou; Bernard Goldfarb; Fionn Murtagh; Catherine Pardoux; Myriam Touati. Statistical learning and data science, Chapman Hall/CRC, pp.135-145, 2012, 978-1-4398-6763-1. ⟨hal-01125864⟩



Record views


Files downloads