ISBN 978-951-22-8929-5 (printed version)

For decades, much research has been devoted to developing and comparing variable selection methods, but primarily for the classical case of independent observations. Existing variable-selection methods can be adapted to cluster-correlated observations, but some adaptation is required. For example, classical model fit statistics such as AIC and BIC are undefined if the likelihood function is unknown (Pan, 2001). Little research has been done on variable selection for generalized estimating equations (GEE, Liang and Zeger, 1986) and similar correlated data approaches. This thesis will review existing work on model selection for GEE and propose new model selection options for GEE, as well as for a more sophisticated marginal modeling approach based on quadratic inference functions (QIF, Qu, Lindsay, and Li, 2000), which has better asymptotic properties than classic GEE. The focus is on selection using continuous penalties such as LASSO (Tibshirani, 1996) or SCAD (Fan and Li, 2001) rather than the older discrete penalties such as AIC and BIC. The

estimates of posterior model probabilities

We consider the specification of prior distributions for Bayesian model comparison, focusing on regression-type models. We propose a particular joint specification of the prior distribution across models so that sensitivity of posterior model probabilities to the dispersion of prior distributions for the parameters of individual models (Lindley’s paradox) is diminished. We illustrate the behavior of inferential and predictive posterior quantities in linear and log-linear regressions under our proposed prior densities with a series of simulated and real data examples.

Traditional criteria for comparing alternative Bayesian hierarchical models, such as cross validation sums of squares, are inappropriate for non-standard data structures. More flexible cross validation criteria such as predictive densities facilitate effective evaluations across a broader range of data structures, but do so at the expense of introducing computational challenges. This paper considers Markov Chain Monte Carlo strategies for calculating Bayesian predictive densities for vector measurements subject to differential component-wise censoring. It discusses computational obstacles in Bayesian computations resulting from both the multivariate and incomplete nature of the data, and suggests two Monte Carlo approaches for implementing predictive density calculations. It illustrates the value of the proposed methods in the context of comparing alternative models for joint distributions of contaminant concentration measurements.

journal homepage: www.elsevier.com/locate/caie A collaborative demand forecasting process with event-based fuzzy judgements

Bayesian estimation in Markov random fields is very hard due to the intractability of the partition function. The introduction of hidden units makes the situation even worse due to the presence of potentially very many modes in the posterior distribution. For the first time we propose a comprehensive procedure to address one of the Bayesian estimation problems, approximating the evidence of partially observed MRFs based on the Laplace approximation. We also introduce a number of approximate MCMC-based methods for comparison but find that the Laplace approximation significantly outperforms these. 1