BAYESIAN SUBSET SELECTION AND MODEL AVERAGING USING A CENTRED AND DISPERSED PRIOR FOR THE ERROR VARIANCE

This article proposes a new data‐based prior distribution for the error variance in a Gaussian linear regression model, when the model is used for Bayesian variable selection and model averaging. For a given subset of variables in the model, this prior has a mode that is an unbiased estimator of the error variance but is suitably dispersed to make it uninformative relative to the marginal likelihood. The advantage of this empirical Bayes prior for the error variance is that it is centred and dispersed sensibly and avoids the arbitrary specification of hyperparameters. The performance of the new prior is compared to that of a prior proposed previously in the literature using several simulated examples and two loss functions. For each example our paper also reports results for the model that orthogonalizes the predictor variables before performing subset selection. A real example is also investigated. The empirical results suggest that for both the simulated and real data, the performance of the estimators based on the prior proposed in our article compares favourably with that of a prior used previously in the literature.     

A hierarchical model for extreme wind speeds

Summary.  A typical extreme value analysis is often carried out on the basis of simplistic inferential procedures, though the data being analysed may be structurally complex. Here we develop a hierarchical model for hourly gust maximum wind speed data, which attempts to identify site and seasonal effects for the marginal densities of hourly maxima, as well as for the serial dependence at each location. A Gaussian model for the random effects exploits the meteorological structure in the data, enabling increased precision for inferences at individual sites and in individual seasons. The Bayesian framework that is adopted is also exploited to obtain predictive return level estimates at each site, which incorporate uncertainty due to model estimation, as well as the randomness that is inherent in the processes that are involved.     

Boosting with Bayesian stumps

Boosting is a new, powerful method for classification. It is an iterative procedure which successively classifies a weighted version of the sample, and then reweights this sample dependent on how successful the classification was. In this paper we review some of the commonly used methods for performing boosting and show how they can be fit into a Bayesian setup at each iteration of the algorithm. We demonstrate how this formulation gives rise to a new splitting criterion when using a domain-partitioning classification method such as a decision tree. Further we can improve the predictive performance of simple decision trees, known as stumps, by using a posterior weighted average of them to classify at each step of the algorithm, rather than just a single stump. The main advantage of this approach is to reduce the number of boosting iterations required to produce a good classifier with only a minimal increase in the computational complexity of the algorithm.     

Nonparametric empirical Bayes for the Dirichlet process mixture model

The Dirichlet process prior allows flexible nonparametric mixture modeling. The number of mixture components is not specified in advance and can grow as new data arrive. However, analyses based on the Dirichlet process prior are sensitive to the choice of the parameters, including an infinite-dimensional distributional parameter G ₀. Most previous applications have either fixed G ₀ as a member of a parametric family or treated G ₀ in a Bayesian fashion, using parametric prior specifications. In contrast, we have developed an adaptive nonparametric method for constructing smooth estimates of G ₀. We combine this method with a technique for estimating α, the other Dirichlet process parameter, that is inspired by an existing characterization of its maximum-likelihood estimator. Together, these estimation procedures yield a flexible empirical Bayes treatment of Dirichlet process mixtures. Such a treatment is useful in situations where smooth point estimates of G ₀ are of intrinsic interest, or where the structure of G ₀ cannot be conveniently modeled with the usual parametric prior families. Analysis of simulated and real-world datasets illustrates the robustness of this approach.     

Defining attributes for scorecard construction in credit scoring

In many domains, simple forms of classification rules are needed because of requirements such as ease of use. A particularly simple form splits each variable into just a few categories, assigns weights to the categories, sums the weights for a new object to be classified, and produces a classification by comparing the score with a threshold. Such instruments are often called scorecards. We describe a way to find the best partition of each variable using a simulated annealing strategy. We present theoretical and empirical comparisons of two such additive models, one based on weights of evidence and another based on logistic regression.     

Accuracy of approximate progressively censored reliability sampling plans for exponential models

Reliability sampling plans provide an efficient method to determine the acceptability of a product based upon the lifelengths of some test units. Usually, they depend on the producer and consumer’s quality requirements and do not admit closed form solutions. Acceptance sampling plans for one- and two-parameter exponential lifetime models, derived by approximating the operating characteristic curve, are presented in this paper. The accuracy of these approximate plans, which are explicitly expressible and valid for failure and progressive censoring, is assessed. The approximation proposed in the one-parameter case is found to be practically exact. Explicit lower and upper bounds on the smallest sample size are given in the two-parameter case. Some additional advantages are also pointed out.