- Bayesian Anal.
- Advance publication (2018), 27 pages.
Bayesian Inference in Nonparanormal Graphical Models
Gaussian graphical models have been used to study intrinsic dependence among several variables, but the Gaussianity assumption may be restrictive in many applications. A nonparanormal graphical model is a semiparametric generalization for continuous variables where it is assumed that the variables follow a Gaussian graphical model only after some unknown smooth monotone transformations on each of them. We consider a Bayesian approach in the nonparanormal graphical model by putting priors on the unknown transformations through a random series based on B-splines where the coefficients are ordered to induce monotonicity. A truncated normal prior leads to partial conjugacy in the model and is useful for posterior simulation using Gibbs sampling. On the underlying precision matrix of the transformed variables, we consider a spike-and-slab prior and use an efficient posterior Gibbs sampling scheme. We use the Bayesian Information Criterion to choose the hyperparameters for the spike-and-slab prior. We present a posterior consistency result on the underlying transformation and the precision matrix. We study the numerical performance of the proposed method through an extensive simulation study and finally apply the proposed method on a real data set.
Bayesian Anal., Advance publication (2018), 27 pages.
First available in Project Euclid: 5 June 2019
Permanent link to this document
Digital Object Identifier
Mulgrave, Jami J.; Ghosal, Subhashis. Bayesian Inference in Nonparanormal Graphical Models. Bayesian Anal., advance publication, 5 June 2019. doi:10.1214/19-BA1159. https://projecteuclid.org/euclid.ba/1559721629
- Supplementary Material for Bayesian Inference in Nonparanormal Graphical Models. The supplement that includes the proof to the consistency theorems. https://github.com/jnj2102/BayesianNonparanormal.
- GitHub Repository: Bayesian Nonparanormal. The code used to run the methods described in this paper are available on GitHub.