Abstract:Statistical inference in high-dimensional settings is challenging when standard unregularized methods are employed. In this work, we focus on the case of multiple correlated proportions for which we develop a Bayesian inference framework. For this purpose, we construct an $m$-dimensional Beta distribution from a $2^m$-dimensional Dirichlet distribution, building on work by Olkin and Trikalinos (2015). This readily leads to a multivariate Beta-binomial model for which simple update rules from the common Dirichlet-multinomial model can be adopted. From the frequentist perspective, this approach amounts to adding pseudo-observations to the data and allows a joint shrinkage estimation of mean vector and covariance matrix. For higher dimensions ($m > 10$), the extensive model based on $2^m$ parameters starts to become numerically infeasible. To counter this problem, we utilize a reduced parametrisation which has only $1 + m(m + 1)/2$ parameters describing first and second order moments. A copula model can then be used to approximate the (posterior) multivariate Beta distribution. A natural inference goal is the construction of multivariate credible regions. The properties of different credible regions are assessed in a simulation study in the context of investigating the accuracy of multiple binary classifiers. It is shown that the extensive and copula approach lead to a (Bayes) coverage probability very close to the target level. In this regard, they outperform credible regions based on a normal approximation of the posterior distribution, in particular for small sample sizes. Additionally, they always lead to credible regions which lie entirely in the parameter space which is not the case when the normal approximation is used.

Correlated binomial models and correlation structures

Correlation Structures of Correlated Binomial Models and Implied Default Distribution

Extended Correlation Order and Some of Its Properties

On the Correlation Order

A Set of Efficient Methods to Generate High-Dimensional Binary Data with Specified Correlation Structures

Correlated Binomial Process

Sums of Possibly Associated Bernoulli Variables: The Conway-Maxwell-Binomial Distribution

Simultaneous Inference for Multiple Proportions: A Multivariate Beta-Binomial Model

Constructions for a bivariate beta distribution

Two Generalizations of the Binomial Distribution

Correlation Models with Long-Range Dependence

Markov counting models for correlated binary responses

Invariant correlation under marginal transforms

Modeling paired binary data by a new bivariate Bernoulli model with flexible beta kernel correlation

Distance Correlation in Multiple Biased Sampling Models

Correlated random networks

Central limit theorem for a class of globally correlated random variables

A mixture distribution for modelling bivariate ordinal data

A class of 2 X 2 correlated random-matrix models with Brody spacing distribution

Weak Poissonian box correlations of higher order

On Conditional Correlations