skip to main content

Find a Project

Back to Project Search

Scalable Bayesian inference for structural equation models

Project

Project Details

Program
Statistics
Field of Study
Statistics / Applied Math
Division
Computer, Electrical and Mathematical Sciences and Engineering
Faculty Lab Link
https://bayescomp.kaust.edu.sa

Project Description

This project aims to equip latent structural equation models with a powerful and highly efficient Bayesian inference engine called INLA (www.r-inla.org), a state-of-the-art computational framework that delivers the speed and scalability of modern machine learning while preserving the interpretability and flexibility of traditional statistical models. Leveraging INLA’s capabilities for structural equation models and the like will unlock a new wave of scalable, domain-ready tools that expand the reach and relevance of latent variable analysis. The public good is substantial: more accurate diagnostics in healthcare, better-targeted interventions in education, and smarter decision-making in digital and data-driven industries.

About the Researcher

Haavard Rue
Professor, Statistics
Computer, Electrical and Mathematical Science and Engineering Division

Affiliations

Education Profile

  • PhD Norwegian Institute of Technology, 1993
  • MEng Norwegian Institute of Technology, 1988

Research Interests

Professor Rue's research interests lie in computational Bayesian statistics and Bayesian methodology such as priors, sensitivity and robustness. His main body of research is built around the R-INLA project (www.r-inla.org), which aims to provide a practical tool for approximate Bayesian analysis of latent Gaussian models, often at extreme data scales. This project also includes efforts to use stochastic partial differential equations to represent Gaussian fields, for the use in spatial statistics.

Selected Publications

  • H. Rue, S. Martino, and N. Chopin. a€œApproximate Bayesian Inference for Latent Gaussian Models Using Integrated Nested Laplace Approximations (with dis-cussion)a€. In: Journal of the Royal Statistical Society, Series B 71.2 (2009), pp. 319a-392.
  • F. Lindgren, H. Rue, and J. LindstrA¶m. a€œAn explicit link between Gaussian fields and Gaussian Markov random fields: The SPDE approach (with discussion)a€. In: Journal of the Royal Statistical Society, Series B 73.4 (2011), pp. 423a-498.
  • D. Simpson, J. Illian, F. Lindgren, S. SA¸rbye, and H. Rue. a€œGoing off grid: Com-putational efficient inference for log-Gaussian Cox processesa€. In: Biometrika 103.1 (2016). (doi: 10.1093/biomet/asv064), pp. 1a-22.
  • D. P. Simpson, H. Rue, T. G. Martins, A. Riebler, and S. H. SA¸rbye. Penalising model component complexity: A principled, practical approach to constructing priors. arXiv:1403.4630 (revised in 2015). NTNU, Trondheim, Norway, 2014.
  • H. Rue and L. Held. Gaussian Markov Random Fields: Theory and Applications. Vol. 104. Monographs on Statistics and Applied Probability. London: Chapman & Hall, 2005.

Desired Project Deliverables

Project deliverables will be determined based on the candidate's background and may include method development, R coding, case studies, and other relevant research tasks.

Recommended Student Background

Statistics
Programming (C/R/Python)
Applied Mathematics
On this page