Recent technological advances have enabled researchers in a variety of fields to collect accurately geocoded data for several variables simultaneously. In many cases it may be most appropriate to jointly model these multivariate spatial processes without constraints on their conditional relationships. When data have been collected on a regular lattice, the multivariate conditionally autoregressive (MCAR) models are a common choice. However, inference from these MCAR models relies heavily on the pre-specified neighborhood structure and often assumes a separable covariance structure. Here, we present a multivariate spatial model using a spectral analysis approach that enables inference on the conditional relationships between the variables that does not rely on a pre-specified neighborhood structure, is non-separable, and is computationally efficient. Covariance and cross-covariance functions are defined in the spectral domain to obtain computational efficiency. The resulting pseudo posterior inference on the correlation matrix allows for quantification of the conditional dependencies. A comparison is made with an MCAR model that is shown to be highly sensitive to the choice of neighborhood. The approaches are illustrated for the toxic element arsenic and four other soil elements whose relative concentrations were measured on a microscale spatial lattice. Understanding conditional relationships between arsenic and other soil elements provides insights for mitigating pervasive arsenic poisoning in drinking water in southern Asia and elsewhere.

Network data are increasingly collected along with other variables of interest. Our motivation is drawn from neurophysiology studies measuring brain connectivity networks for a sample of individuals along with their membership to a low or high creative reasoning group. It is of paramount importance to develop statistical methods for testing of global and local changes in the structural interconnections among brain regions across groups. We develop a general Bayesian procedure for inference and testing of group differences in the network structure, which relies on a nonparametric representation for the conditional probability mass function associated with a network-valued random variable. By leveraging a mixture of low-rank factorizations, we allow simple global and local hypothesis testing adjusting for multiplicity. An efficient Gibbs sampler is defined for posterior computation. We provide theoretical results on the flexibility of the model and assess testing performance in simulations. The approach is applied to provide novel insights on the relationships between human brain networks and creativity.

In the Bayesian framework a standard approach to model criticism is to compare some function of the observed data to a reference predictive distribution. The result of the comparison can be summarized in the form of a $p$-value, and computation of some kinds of Bayesian predictive $p$-values can be challenging. The use of regression adjustment approximate Bayesian computation (ABC) methods is explored for this task. Two problems are considered. The first is approximation of distributions of prior predictive $p$-values for the purpose of choosing weakly informative priors in the case where the model checking statistic is expensive to compute. Here the computation is difficult because of the need to repeatedly sample from a prior predictive distribution for different values of a prior hyperparameter. The second problem considered is the calibration of posterior predictive $p$-values so that they are uniformly distributed under some reference distribution for the data. Computation is difficult because the calibration process requires repeated approximation of the posterior for different data sets under the reference distribution. In both these problems we argue that high accuracy in the computations is not required, which makes fast approximations such as regression adjustment ABC very useful. We illustrate our methods with several examples.

In difficult object segmentation tasks, utilizing image information alone is not sufficient; incorporation of object shape prior models is necessary to obtain competitive segmentation performance. Most formulations that incorporate both shape and image information are in the form of energy functional optimization problems. This paper introduces a Bayesian latent marked Poisson process for segmenting multiple objects in an image. The model takes both shape and image feature/appearance into account—it generates object locations from a spatial Poisson process, then generates shape parameters from a shape prior model as the latent marks. Inferentially, this partitions the image: pixels inside objects are assumed to be generated from an object observation/appearance model and pixels outside objects come from a background model. The Poisson process provides (non-homogeneous) spatial priors for object locations and the marks allow the incorporation of shape priors. We develop a hybrid Gibbs sampler that addresses the variation in model order and nonconjugacy that arise in this setting and we present experimental results on synthetic images and two diverse domains in real images: cell segmentation in biological images and pedestrian and car detection in traffic images.

Bayesian item response models have been used in modeling educational testing and Internet ratings data. Typically, the statistical analysis is carried out using Markov Chain Monte Carlo methods. However, these may not be computationally feasible when real-time data continuously arrive and online parameter estimation is needed. We develop an efficient algorithm based on a deterministic moment-matching method to adjust the parameters in real-time. The proposed online algorithm works well for two real datasets, achieving good accuracy but with considerably less computational time.

Optimal experimental design is an important methodology for most efficiently allocating resources in an experiment to best achieve some goal. Bayesian experimental design considers the potential impact that various choices of the controllable variables have on the posterior distribution of the unknowns. Optimal Bayesian design involves maximising an expected utility function, which is an analytically intractable integral over the prior predictive distribution. These integrals are typically estimated via standard Monte Carlo methods. In this paper, we demonstrate that the use of randomised quasi-Monte Carlo can bring significant reductions to the variance of the estimated expected utility. This variance reduction can then lead to a more efficient optimisation algorithm for maximising the expected utility.

This paper investigates the use of regularization priors in the context of treatment effect estimation using observational data where the number of control variables is large relative to the number of observations. First, the phenomenon of “regularization-induced confounding” is introduced, which refers to the tendency of regularization priors to adversely bias treatment effect estimates by over-shrinking control variable regression coefficients. Then, a simultaneous regression model is presented which permits regularization priors to be specified in a way that avoids this unintentional “re-confounding”. The new model is illustrated on synthetic and empirical data.

We present a Bayesian model for estimating the joint distribution of multivariate categorical data when units are nested within groups. Such data arise frequently in social science settings, for example, people living in households. The model assumes that (i) each group is a member of a group-level latent class, and (ii) each unit is a member of a unit-level latent class nested within its group-level latent class. This structure allows the model to capture dependence among units in the same group. It also facilitates simultaneous modeling of variables at both group and unit levels. We develop a version of the model that assigns zero probability to groups and units with physically impossible combinations of variables. We apply the model to estimate multivariate relationships in a subset of the American Community Survey. Using the estimated model, we generate synthetic household data that could be disseminated as redacted public use files. Supplementary materials (Hu et al., 2017) for this article are available online.

In contingency table analysis, sparse data is frequently encountered for even modest numbers of variables, resulting in non-existence of maximum likelihood estimates. A common solution is to obtain regularized estimates of the parameters of a log-linear model. Bayesian methods provide a coherent approach to regularization, but are often computationally intensive. Conjugate priors ease computational demands, but the conjugate Diaconis–Ylvisaker priors for the parameters of log-linear models do not give rise to closed form credible regions, complicating posterior inference. Here we derive the optimal Gaussian approximation to the posterior for log-linear models with Diaconis–Ylvisaker priors, and provide convergence rate and finite-sample bounds for the Kullback–Leibler divergence between the exact posterior and the optimal Gaussian approximation. We demonstrate empirically in simulations and a real data application that the approximation is highly accurate, even for modest sample sizes. We also propose a method for model selection using the approximation. The proposed approximation provides a computationally scalable approach to regularized estimation and approximate Bayesian inference for log-linear models.

We present a locally adaptive nonparametric curve fitting method that operates within a fully Bayesian framework. This method uses shrinkage priors to induce sparsity in order-$k$ differences in the latent trend function, providing a combination of local adaptation and global control. Using a scale mixture of normals representation of shrinkage priors, we make explicit connections between our method and $k$th order Gaussian Markov random field smoothing. We call the resulting processes shrinkage prior Markov random fields (SPMRFs). We use Hamiltonian Monte Carlo to approximate the posterior distribution of model parameters because this method provides superior performance in the presence of the high dimensionality and strong parameter correlations exhibited by our models. We compare the performance of three prior formulations using simulated data and find the horseshoe prior provides the best compromise between bias and precision. We apply SPMRF models to two benchmark data examples frequently used to test nonparametric methods. We find that this method is flexible enough to accommodate a variety of data generating models and offers the adaptive properties and computational tractability to make it a useful addition to the Bayesian nonparametric toolbox.

We introduce a computationally efficient Bayesian model for predicting high-dimensional dependent count-valued data. In this setting, the Poisson data model with a latent Gaussian process model has become the de facto model. However, this model can be difficult to use in high dimensional settings, where the data may be tabulated over different variables, geographic regions, and times. These computational difficulties are further exacerbated by acknowledging that count-valued data are naturally non-Gaussian. Thus, many of the current approaches, in Bayesian inference, require one to carefully calibrate a Markov chain Monte Carlo (MCMC) technique. We avoid MCMC methods that require tuning by developing a new conjugate multivariate distribution. Specifically, we introduce a multivariate log-gamma distribution and provide substantial methodological development of independent interest including: results regarding conditional distributions, marginal distributions, an asymptotic relationship with the multivariate normal distribution, and full-conditional distributions for a Gibbs sampler. To incorporate dependence between variables, regions, and time points, a multivariate spatio-temporal mixed effects model (MSTM) is used. To demonstrate our methodology we use data obtained from the US Census Bureau’s Longitudinal Employer-Household Dynamics (LEHD) program. In particular, our approach is motivated by the LEHD’s Quarterly Workforce Indicators (QWIs), which constitute current estimates of important US economic variables.