Order selection in finite mixtures of linear regressions

Finite mixture models can adequately model population heterogeneity when this heterogeneity arises from a finite number of relatively homogeneous clusters. An example of such a situation is market segmentation. Order selection in mixture models, i.e. selecting the correct number of components, however, is a problem which has not been satisfactorily resolved. Existing simulation results in the literature do not completely agree with each other. Moreover, it appears that the performance of different selection methods is affected by the type of model and the parameter values. Furthermore, most existing results are based on simulations where the true generating model is identical to one of the models in the candidate set. In order to partly fill this gap we carried out a (relatively) large simulation study for finite mixture models of normal linear regressions. We included several types of model (mis)specification to study the robustness of 18 order selection methods. Furthermore, we compared the performance of these selection methods based on unpenalized and penalized estimates of the model parameters. The results indicate that order selection based on penalized estimates greatly improves the success rates of all order selection methods. The most successful methods were \(MDL2\) , \(MRC\) , \(MRC_k\) , \(ICL\) – \(BIC\) , \(ICL\) , \(CAIC\) , \(BIC\) and \(CLC\) but not one method was consistently good or best for all types of model (mis)specification.     

An Automated MCEM Algorithm for Hierarchical Models with Multivariate and Multitype Response Variables

In this article, we consider a model allowing the analysis of multivariate data, which can contain data attributes of different types (e.g., continuous, discrete, binary). This model is a two-level hierarchical model which supports a wide range of correlation structures and can accommodate overdispersed data. Maximum likelihood estimation of the model parameters is achieved with an automated Monte Carlo expectation maximization algorithm. Our method is tested in a simulation study in the bivariate case and applied to a data set dealing with beehive activity.     

Public vs. Private Offers in the Market for Lemons

We study the role of observability in bargaining with correlated values. Short‐run buyers sequentially submit offers to one seller. When previous offers are observable, bargaining is likely to end up in an impasse. In contrast, when offers are hidden, agreement is always reached, although with delay.     

Perfect sampling algorithm for small <Emphasis Type="Italic">m</Emphasis>×<Emphasis Type="Italic">n</Emphasis> contingency tables

A Markov chain is proposed that uses coupling from the past sampling algorithm for sampling m×n contingency tables. This method is an extension of the one proposed by Kijima and Matsui (Rand. Struct. Alg., 29:243–256, 2006). It is not polynomial, as it is based upon a recursion, and includes a rejection phase but can be used for practical purposes on small contingency tables as illustrated in a classical 4×4 example.     

The Environmental Geochemistry and Bioaccessibility of Mercury in Soils and Sediments: A Review

There have been many studies of mercury geochemistry in the environment and its bioconcentration/bioaccumulation through the aquatic food chain. However, there is a dearth of information regarding the bioaccessibility of mercury in human receptors exposed primarily by soil ingestion. This paper reviews the current state of knowledge of mercury bioaccessibility and speciation in soils, and the utility of speciation methods to estimate mercury bioaccessibility. We conclude that additional research is necessary to determine: (1) whether analytical measurements can adequately determine the bioaccessibility of mercury in sediments and soils; (2) the accuracy of in vitro analyses in assessing mercury bioaccessibility; (3) the ability of mercury to cross tissue membranes of the mouth, esophagus, stomach, and the small and large intestines; (4) the speciation and distribution of mercury in biological fluids; and (5) mercury bioavailability using an in vivo animal model relevant to human gastrointestinal tract conditions.     

Effects of sleep inertia on cognitive performance following a 1-hour nap

This study investigated the effects of a 1-h nap on subsequent performance in spatial memory (SM) and logical reasoning (LR) tasks. The objective was to evaluate the duration of the effects of sleep inertia (SI). The performance was measured in two independent groups of subjects. The 1-h nap took place at 00:00 h in group 1 and at 03:00 h in group 2. In each task, the experimental design comprised one no-nap condition in which subjects had no sleep before the night tests, and a nap condition that comprised the 1-h nap followed by the test sessions. To measure the duration of SI effects, the subjects were tested in two 30-min sessions and the data in each session were analysed in sub-units of time of 3 min each. In each task the results showed no effects on accuracy, and no circadian effects of napping were found. In each task, analyses of pooled data of the two groups showed that the performance in the 1-h nap condition exhibited significant reductions of speed immediately following awakening, when compared with no-nap, reflecting SI effects. In SM, the slowing lasted 24 min, and 27 min in LR. Practical implications of the results are discussed.     

Decision-Making Under Scientific Uncertainty: The Economics of the Precautionary Principle

The Precautionary Principle has provided the foundations for building a new risk regulatory pattern under scientific uncertainty. This paper investigates how classical economic theory may, or may not, justify the Precautionary Principle. It examines the link between irreversibility, the prospect of increasing information over time and risk management. In doing so, it brings closer the notion of option value to that of precaution. Using a general modelling framework, it identifies the conditions so that the Precautionary Principle is an efficient economic guideline. It also explains why precautionary policies are not likely to emerge in a competitive economy or in the presence of a global pollution problem.     

Socio‐Emotional Context and Adolescents' Decision Making: The Experience of Regret and Relief After Social Comparison

Many studies indicate that adolescents are hypersensitive to rewards in salient socio‐emotional contexts. However, little is known about adolescents' sensitivity to complex negative emotions. To study the development of the ability to experience regret and relief, children, adolescents, and young adults performed a gambling task in a socio‐emotional context of competition, in which they were informed that their outcome would be compared with that of a competitor. The context of competition impacted the feelings of regret and relief in adolescents. When adolescents experienced an initial negative outcome, their feeling of relief was reinforced. However, they did not seem to experience regret in the social context of competition. These results provide a possible explanation for adolescents' enhanced risk seeking.     

INFORMATION THROUGH SAMPLING FROM A BINOMIAL DISTRIBUTION*

Lindley's measure of experimental information is utilized to determine the optimal sample-size that is required to obtain a prescribed level of accuracy about the parameter of a binomial distribution. The measure is expressed through its approximate limiting relationship to the posterior variance of the parameter, and a simple application of the methodology is presented.