Intuitive joint priors for variance parameters
From MaRDI portal
Abstract: Variance parameters in additive models are typically assigned independent priors that do not account for model structure. We present a new framework for prior selection based on a hierarchical decomposition of the total variance along a tree structure to the individual model components. For each split in the tree, an analyst may be ignorant or have a sound intuition on how to attribute variance to the branches. In the former case a Dirichlet prior is appropriate to use, while in the latter case a penalised complexity (PC) prior provides robust shrinkage. A bottom-up combination of the conditional priors results in a proper joint prior. We suggest default values for the hyperparameters and offer intuitive statements for eliciting the hyperparameters based on expert knowledge. The prior framework is applicable for R packages for Bayesian inference such as INLA and RStan. Three simulation studies show that, in terms of the application-specific measures of interest, PC priors improve inference over Dirichlet priors when used to penalise different levels of complexity in splits. However, when expressing ignorance in a split, Dirichlet priors perform equally well and are preferred for their simplicity. We find that assigning current state-of-the-art default priors for each variance parameter individually is less transparent and does not perform better than using the proposed joint priors. We demonstrate practical use of the new framework by analysing spatial heterogeneity in neonatal mortality in Kenya in 2010-2014 based on complex survey data.
Recommendations
- Joint prior distributions for variance parameters in Bayesian analysis of normal hierarchical models
- Scale-dependent priors for variance parameters in structured additive distributional regression
- Conservative prior distributions for variance parameters in hierarchical models
- Prior distributions for variance parameters in hierarchical models (Comment on article by Browne and Draper)
- A prior for the variance in hierarchical models
Cites work
- scientific article; zbMATH DE number 621787 (Why is no real title available?)
- A Primer on Statistical Distributions
- A general modelling framework for multivariate disease mapping
- A weakly informative default prior distribution for logistic and other regression models
- Advanced spatial modeling with stochastic partial differential equations using R and INLA
- An explicit link between Gaussian fields and Gaussian Markov random fields: the stochastic partial differential equation approach
- Approximate Bayesian inference for latent Gaussian models by using integrated nested Laplace approximations (with discussion)
- Bayesian data analysis.
- Bayesian image restoration, with two applications in spatial statistics (with discussion)
- Constructing Priors that Penalize the Complexity of Gaussian Random Fields
- Dirichlet-Laplace priors for optimal shrinkage
- Disease mapping and spatial regression with count data
- Gaussian Markov Random Fields
- Hierarchical modeling and analysis for spatial data
- Penalised complexity priors for stationary autoregressive processes
- Penalising model component complexity: a principled, practical approach to constructing priors
- Prior distributions for variance parameters in hierarchical models (Comment on article by Browne and Draper)
- Spatial and spatio-temporal Bayesian models with R-INLA
- Strictly Proper Scoring Rules, Prediction, and Estimation
Cited in
(8)- The R2D2 Prior for Generalized Linear Mixed Models
- Prior knowledge elicitation: the past, present, and future
- Bayesian multiresolution modeling of georeferenced data: an extension of `LatticeKrig'
- A standardization procedure to incorporate variance partitioning-based priors in latent Gaussian models
- Design and structure dependent priors for scale parameters in latent Gaussian models
- Asymptotics for a class of parametric martingale posteriors
- A variance partitioning multi-level model for forest inventory data with a fixed plot design
- Scale-dependent priors for variance parameters in structured additive distributional regression
Describes a project that uses
Uses Software
This page was built for publication: Intuitive joint priors for variance parameters
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2057345)
