Spatial point pattern data describes locations of events observed over a given domain, with the number of and locations of these events being random. Historically, data analysis for spatial point patterns has focused on rejecting complete spatial randomness and then on fitting a richer model specification. From a Bayesian standpoint, the literature is growing but primarily considers versions of Poisson processes, focusing on specifications for the intensity. However, the Bayesian literature on, e.g., clustering or inhibition processes is limited, primarily attending to model fitting. There is little attention given to full inference and scant with regard to model adequacy or model comparison.

The contribution here is full Bayesian analysis, implemented through generation of posterior point patterns using composition. Model features, hence broad inference, can be explored through functions of these samples. The approach is general, applicable to any generative model for spatial point patterns.

The approach is also useful in considering model criticism and model selection both in-sample and, when possible, out-of-sample. Here, we adapt or extend familiar tools. In particular, for model criticism, we consider Bayesian residuals, realized and predictive, along with empirical coverage and prior predictive checks through Monte Carlo tests. For model choice, we propose strategies using predictive mean square error, empirical coverage, and ranked probability scores. For simplicity, we illustrate these methods with standard models such as Poisson processes, log-Gaussian Cox processes, and Gibbs processes. The utility of our approach is demonstrated using a simulation study and two real datasets.

Conventional phase II clinical trials use either a single-arm or a double-arm scheme to examine the treatment effect of an investigational drug. The hypotheses tests under these two schemes are different, as a single-arm study usually tests the response rate of the new drug against a set of fixed reference rates and a double-arm randomized trial compares the new drug with the standard treatment or placebo. To bridge the single- and double-arm schemes in one phase II clinical trial, we propose a Bayesian two-stage design with changing hypothesis tests. Stage 1 enrolls patients solely to the experimental arm to make a comparison with the reference rates, and stage 2 imposes a double-arm comparison of the experimental arm with the control arm. The design is calibrated with respect to error rates from both the frequentist and Bayesian perspectives. Moreover, we control the “type III error rate”, defined as the probability of prematurely stopping the trial at stage 1 when the trial is supposed to move on to stage 2. We conduct extensive simulations on the calculations of these error rates to examine the operational characteristics of our proposed method, and illustrate it with a non-small cell lung cancer trial.

We provide sufficient conditions to derive posterior concentration rates for Aalen counting processes on a finite time horizon. The conditions are designed to resemble those proposed in the literature for the problem of density estimation, for instance, in Ghosal et al. (2000), so that existing results on density estimation can be adapted to the present setting. We apply the general theorem to some prior models including Dirichlet process mixtures of uniform densities to estimate monotone nondecreasing intensities and log-splines.

We study the problem of independence and conditional independence tests between categorical covariates and a continuous response variable, which has an immediate application in genetics. Instead of estimating the conditional distribution of the response given values of covariates, we model the conditional distribution of covariates given the discretized response (aka “slices”). By assigning a prior probability to each possible discretization scheme, we can compute efficiently a Bayes factor (BF)-statistic for the independence (or conditional independence) test using a dynamic programming algorithm. Asymptotic and finite-sample properties such as power and null distribution of the BF statistic are studied, and a stepwise variable selection method based on the BF statistic is further developed. We compare the BF statistic with some existing classical methods and demonstrate its statistical power through extensive simulation studies. We apply the proposed method to a mouse genetics data set aiming to detect quantitative trait loci (QTLs) and obtain promising results.

The general projected normal distribution is a simple and intuitive model for directional data in any dimension: a multivariate normal random vector divided by its length is the projection of that vector onto the surface of the unit hypersphere. Observed data consist of the projections, but not the lengths. Inference for this model has been restricted to the two-dimensional (circular) case, using Bayesian methods with data augmentation to generate the latent lengths and a Metropolis-within-Gibbs algorithm to sample from the posterior. We describe a new parameterization of the general projected normal distribution that makes inference in any dimension tractable, including the important three-dimensional (spherical) case, which has not previously been considered. Under this new parameterization, the full conditionals of the unknown parameters have closed forms, and we propose a new slice sampler to draw the latent lengths without the need for rejection. Gibbs sampling with this new scheme is fast and easy, leading to improved Bayesian inference; for example, it is now feasible to conduct model selection among complex mixture and regression models for large data sets. Our parameterization also allows straightforward incorporation of covariates into the covariance matrix of the multivariate normal, increasing the ability of the model to explain directional data as a function of independent regressors. Circular and spherical cases are considered in detail and illustrated with scientific applications. For the circular case, seasonal variation in time-of-day departures of anglers from recreational fishing sites is modeled using covariates in both the mean vector and covariance matrix. For the spherical case, we consider paired angles that describe the relative positions of carbon atoms along the backbone chain of a protein. We fit mixtures of general projected normals to these data, with the best-fitting mixture accurately describing biologically meaningful structures including helices, $\beta$-sheets, and coils and turns. Finally, we show via simulation that our methodology has satisfactory performance in some 10-dimensional and 50-dimensional problems.

In some linear models, such as those with interactions, it is natural to include the relationship between the regression coefficients in the analysis. In this paper, we consider how robust hierarchical continuous prior distributions can be used to express dependence between the size but not the sign of the regression coefficients. For example, to include ideas of heredity in the analysis of linear models with interactions. We develop a simple method for controlling the shrinkage of regression effects to zero at different levels of the hierarchy by considering the behaviour of the continuous prior at zero. Applications to linear models with interactions and generalized additive models are used as illustrations.

This study proposes $p$-th Tobit quantile regression models with endogenous variables. In the first stage regression of the endogenous variable on the exogenous variables, the assumption that the $\alpha$-th quantile of the error term is zero is introduced. Then, the residual of this regression model is included in the $p$-th quantile regression model in such a way that the $p$-th conditional quantile of the new error term is zero. The error distribution of the first stage regression is modelled around the zero $\alpha$-th quantile assumption by using parametric and semiparametric approaches. Since the value of $\alpha$ is a priori unknown, it is treated as an additional parameter and is estimated from the data. The proposed models are then demonstrated by using simulated data and real data on the labour supply of married women.

We present a novel Bayesian approach to analysing multiple time-series with the aim of detecting abnormal regions. These are regions where the properties of the data change from some normal or baseline behaviour. We allow for the possibility that such changes will only be present in a, potentially small, subset of the time-series. We develop a general model for this problem, and show how it is possible to accurately and efficiently perform Bayesian inference, based upon recursions that enable independent sampling from the posterior distribution. A motivating application for this problem comes from detecting copy number variation (CNVs), using data from multiple individuals. Pooling information across individuals can increase the power of detecting CNVs, but often a specific CNV will only be present in a small subset of the individuals. We evaluate the Bayesian method on both simulated and real CNV data, and give evidence that this approach is more accurate than a recently proposed method for analysing such data.

In this paper we develop and study adaptive empirical Bayesian smoothing splines. These are smoothing splines with both smoothing parameter and penalty order determined via the empirical Bayes method from the marginal likelihood of the model. The selected order and smoothing parameter are used to construct adaptive credible sets with good frequentist coverage for the underlying regression function. We use these credible sets as a proxy to show the superior performance of adaptive empirical Bayesian smoothing splines compared to frequentist smoothing splines.

Multivariate disease mapping enriches traditional disease mapping studies by analysing several diseases jointly. This yields improved estimates of the geographical distribution of risk from the diseases by enabling borrowing of information across diseases. Beyond multivariate smoothing for several diseases, several other variables, such as sex, age group, race, time period, and so on, could also be jointly considered to derive multivariate estimates. The resulting multivariate structures should induce an appropriate covariance model for the data. In this paper, we introduce a formal framework for the analysis of multivariate data arising from the combination of more than two variables (geographical units and at least two more variables), what we have called Multidimensional Disease Mapping. We develop a theoretical framework containing both separable and non-separable dependence structures and illustrate its performance on the study of real mortality data in Comunitat Valenciana (Spain).

In this paper we propose a conceptually straightforward method to estimate the marginal data density value (also called the marginal likelihood). We show that the marginal likelihood is equal to the prior mean of the conditional density of the data given the vector of parameters restricted to a certain subset of the parameter space, $A$, times the reciprocal of the posterior probability of the subset $A$. This identity motivates one to use Arithmetic Mean estimator based on simulation from the prior distribution restricted to any (but reasonable) subset of the space of parameters. By trimming this space, regions of relatively low likelihood are removed, and thereby the efficiency of the Arithmetic Mean estimator is improved. We show that the adjusted Arithmetic Mean estimator is unbiased and consistent.

Approximate Bayesian computation performs approximate inference for models where likelihood computations are expensive or impossible. Instead simulations from the model are performed for various parameter values and accepted if they are close enough to the observations. There has been much progress on deciding which summary statistics of the data should be used to judge closeness, but less work on how to weight them. Typically weights are chosen at the start of the algorithm which normalise the summary statistics to vary on similar scales. However these may not be appropriate in iterative ABC algorithms, where the distribution from which the parameters are proposed is updated. This can substantially alter the resulting distribution of summary statistics, so that different weights are needed for normalisation. This paper presents two iterative ABC algorithms which adaptively update their weights and demonstrates improved results on test applications.