We investigate a class of feature allocation models that generalize the Indian buffet process and are parameterized by Gibbs-type random measures. Two existing classes are contained as special cases: the original two-parameter Indian buffet process, corresponding to the Dirichlet process, and the stable (or three-parameter) Indian buffet process, corresponding to the Pitman–Yor process. Asymptotic behavior of the Gibbs-type partitions, such as power laws holding for the number of latent clusters, translates into analogous characteristics for this class of Gibbs-type feature allocation models. Despite containing several different distinct subclasses, the properties of Gibbs-type partitions allow us to develop a black-box procedure for posterior inference within any subclass of models. Through numerical experiments, we compare and contrast a few of these subclasses and highlight the utility of varying power-law behaviors in the latent features.

Heterogeneous networks are useful for modeling complex systems that consist of different types of objects. However, there are limited statistical models to deal with heterogeneous networks. In this paper, we propose a statistical model for community detection in heterogeneous networks. We formulate a heterogeneous version of the mixed membership stochastic blockmodel to accommodate heterogeneity in the data and the content dependent property of the pairwise relationship. We also apply a variational algorithm for posterior inference. The proposed procedure is shown to be consistent for community detection under mixed membership stochastic blockmodels for heterogeneous networks. We demonstrate the advantage of the proposed method in modeling overlapping communities and multiple memberships through simulation studies and applications to a real data set.

Assuming a banded structure is one of the common practice in the estimation of high-dimensional precision matrices. In this case, estimating the bandwidth of the precision matrix is a crucial initial step for subsequent analysis. Although there exist some consistent frequentist tests for the bandwidth parameter, bandwidth selection consistency for precision matrices has not been established in a Bayesian framework. In this paper, we propose a prior distribution tailored to the bandwidth estimation of high-dimensional precision matrices. The banded structure is imposed via the Cholesky factor from the modified Cholesky decomposition. We establish strong model selection consistency for the bandwidth as well as consistency of the Bayes factor. The convergence rates for Bayes factors under both the null and alternative hypotheses are derived which yield similar order of rates. As a by-product, we also propose an estimation procedure for the Cholesky factors yielding an almost optimal order of convergence rates. Two-sample bandwidth test is also considered, and it turns out that our method is able to consistently detect the equality of bandwidths between two precision matrices. The simulation study confirms that our method in general outperforms the existing frequentist and Bayesian methods.

For many biomedical, environmental, and economic studies, the single index model provides a practical dimension reaction as well as a good physical interpretation of the unknown nonlinear relationship between the response and its multiple predictors. However, widespread uses of existing Bayesian analysis for such models are lacking in practice due to some major impediments, including slow mixing of the Markov Chain Monte Carlo (MCMC), the inability to deal with missing covariates and a lack of theoretical justification of the rate of convergence of Bayesian estimates. We present a new Bayesian single index model with an associated MCMC algorithm that incorporates an efficient Metropolis–Hastings (MH) step for the conditional distribution of the index vector. Our method leads to a model with good interpretations and prediction, implementable Bayesian inference, fast convergence of the MCMC and a first-time extension to accommodate missing covariates. We also obtain, for the first time, the set of sufficient conditions for obtaining the optimal rate of posterior convergence of the overall regression function. We illustrate the practical advantages of our method and computational tool via reanalysis of an environmental study.

Learning dependence relationships among variables of mixed types provides insights in a variety of scientific settings and is a well-studied problem in statistics. Existing methods, however, typically rely on copious, high quality data to accurately learn associations. In this paper, we develop a method for scientific settings where learning dependence structure is essential, but data are sparse and have a high fraction of missing values. Specifically, our work is motivated by survey-based cause of death assessments known as verbal autopsies (VAs). We propose a Bayesian approach to characterize dependence relationships using a latent Gaussian graphical model that incorporates informative priors on the marginal distributions of the variables. We demonstrate such information can improve estimation of the dependence structure, especially in settings with little training data. We show that our method can be integrated into existing probabilistic cause-of-death assignment algorithms and improves model performance while recovering dependence patterns between symptoms that can inform efficient questionnaire design in future data collection.

This paper introduces a general class of hierarchical nonparametric prior distributions which includes new hierarchical mixture priors such as the hierarchical Gnedin measures, and other well-known prior distributions such as the hierarchical Pitman-Yor and the hierarchical normalized random measures. The random probability measures are constructed by a hierarchy of generalized species sampling processes with possibly non-diffuse base measures. The proposed framework provides a probabilistic foundation for hierarchical random measures, and allows for studying their properties under the alternative assumptions of diffuse, atomic and mixed base measure. We show that hierarchical species sampling models have a Chinese Restaurants Franchise representation and can be used as prior distributions to undertake Bayesian nonparametric inference. We provide a general sampling method for posterior approximation which easily accounts for non-diffuse base measures such as spike-and-slab.

Infectious diseases such as avian influenza pose a global threat to human health. Mathematical and statistical models can provide key insights into the mechanisms that underlie the spread and persistence of infectious diseases, though their utility is linked to the ability to adequately calibrate these models to observed data. Performing robust inference for these systems is challenging. The fact that the underlying models exhibit complex non-linear dynamics, coupled with practical constraints to observing key epidemiological events such as transmission, requires the use of inference techniques that are able to numerically integrate over multiple hidden states and/or infer missing information. Simulation-based inference techniques such as Approximate Bayesian Computation (ABC) have shown great promise in this area, since they rely on the development of suitable simulation models, which are often easier to code and generalise than routines that require evaluations of an intractable likelihood function. In this manuscript we make some contributions towards improving the efficiency of ABC-based particle Markov chain Monte Carlo methods, and show the utility of these approaches for performing both model inference and model comparison in a Bayesian framework. We illustrate these approaches on both simulated data, as well as real data from an experimental transmission study of highly pathogenic avian influenza in genetically modified chickens.

Directional data emerges in a wide array of applications, ranging from atmospheric sciences to medical imaging. Modeling such data, however, poses unique challenges by virtue of their being constrained to non-Euclidean spaces like manifolds. Here, we present a unified Bayesian framework for inference on the Stiefel manifold using the Matrix Langevin distribution. Specifically, we propose a novel family of conjugate priors and establish a number of theoretical properties relevant to statistical inference. Conjugacy enables translation of these properties to their corresponding posteriors, which we exploit to develop the posterior inference scheme. For the implementation of the posterior computation, including the posterior sampling, we adopt a novel computational procedure for evaluating the hypergeometric function of matrix arguments that appears as normalization constants in the relevant densities.

In many applications with high dimensional covariates, the covariates are naturally structured into different groups which can be used to perform efficient statistical inference. We propose a Bayesian hierarchical model with a spike and slab prior specification to perform group selection in high dimensional linear regression models. While several penalization methods and more recently, some Bayesian approaches are proposed for group selection, theoretical properties of Bayesian approaches have not been studied extensively. In this paper, we provide novel theoretical results for group selection consistency under spike and slab priors which demonstrate that the proposed Bayesian approach has advantages compared to penalization approaches. Our theoretical results accommodate flexible conditions on the design matrix and can be applied to commonly used statistical models such as nonparametric additive models for which very limited theoretical results are available for the Bayesian methods. A shotgun stochastic search algorithm is adopted for the implementation of our proposed approach. We illustrate through simulation studies that the proposed method has better performance for group selection compared to a variety of existing methods.

Factor-analytic Gaussian mixtures are often employed as a model-based approach to clustering high-dimensional data. Typically, the numbers of clusters and latent factors must be fixed in advance of model fitting. The pair which optimises some model selection criterion is then chosen. For computational reasons, having the number of factors differ across clusters is rarely considered.

Here the infinite mixture of infinite factor analysers (IMIFA) model is introduced. IMIFA employs a Pitman-Yor process prior to facilitate automatic inference of the number of clusters using the stick-breaking construction and a slice sampler. Automatic inference of the cluster-specific numbers of factors is achieved using multiplicative gamma process shrinkage priors and an adaptive Gibbs sampler. IMIFA is presented as the flagship of a family of factor-analytic mixtures.

Applications to benchmark data, metabolomic spectral data, and a handwritten digit example illustrate the IMIFA model’s advantageous features. These include obviating the need for model selection criteria, reducing the computational burden associated with the search of the model space, improving clustering performance by allowing cluster-specific numbers of factors, and uncertainty quantification.

This paper presents a novel nonlinear regression model for estimating heterogeneous treatment effects, geared specifically towards situations with small effect sizes, heterogeneous effects, and strong confounding by observables. Standard nonlinear regression models, which may work quite well for prediction, have two notable weaknesses when used to estimate heterogeneous treatment effects. First, they can yield badly biased estimates of treatment effects when fit to data with strong confounding. The Bayesian causal forest model presented in this paper avoids this problem by directly incorporating an estimate of the propensity function in the specification of the response model, implicitly inducing a covariate-dependent prior on the regression function. Second, standard approaches to response surface modeling do not provide adequate control over the strength of regularization over effect heterogeneity. The Bayesian causal forest model permits treatment effect heterogeneity to be regularized separately from the prognostic effect of control variables, making it possible to informatively “shrink to homogeneity”. While we focus on observational data, our methods are equally useful for inferring heterogeneous treatment effects from randomized controlled experiments where careful regularization is somewhat less complicated but no less important. We illustrate these benefits via the reanalysis of an observational study assessing the causal effects of smoking on medical expenditures as well as extensive simulation studies.