Simple Marginally Noninformative Prior Distributions for Covariance Matrices
By A. Huang, Matt P. Wand. 2013
251 citations recorded by the source corpus.
A family of prior distributions for covariance matrices is studied. Members of the family possess the attractive property of all standard deviation and correlation parameters being marginally noninformative for particular hyper-parameter choices. Moreover, the family is quite simple and, for approximate Bayesian inference techniques such as Markov chain Monte Carlo and mean eld variational Bayes, has tractability on par with the Inverse-Wishart conjugate fam-ily of prior distributions. A simulation study shows that the new prior distributions can lead to more accurate sparse covariance matrix estimation.
Subfields: Statistics and Probability
View Paper: https://www.semanticscholar.org/paper/3b118b29d7f1498a8e12051d0d88e7c3403aa48e
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Statistics and Probability
- Andrea Riebler
- Daniel P. Simpson
- Håvard Rue
- Linda S. L. Tan
- David J. Nott
- Sigrunn Holbek Sørbye
- Thiago G. Martins
- Kok-Kwang Phoon
Citations in the corpus
- A Stochastic Variational Framework for Fitting and Diagnosing Generalized Linear Mixed Models (2012, methodology): “tions for R^ is given in Kass and Natarajan (2006). A similar prior was used in Overstall and Forster (2010). Alternatively, one may consider marginally noninformative priors for covariance matrices (Huang and Wand, 2013). Methods in this paper can be extended to these priors easily. 3.1 A partially”
- Penalising Model Component Complexity: A Principled, Practical Approach to Constructing Priors (2014, methodology): “Inverse Wishart distributions, which have convenient conjugacy properties and several extensions to these models that remove some of their basic flaws have been proposed (O’Malley and Zaslavsky, 2008; Huang and Wand, 2013). An alternate method is to note that every covariance matrix can be decomposed”
- Quasi-site-specific multivariate probability distribution model for sparse, incomplete, and three-dimensional spatially varying soil data (2021, methodology): “Phoon (2020), Ching (2020), and Ching and Phoon (2020a) used the”