Maternal and Child Anxiety: Do Attachment Beliefs or Children's Perceptions of Maternal Control Mediate Their Association?

This study tested a model of the association between maternal and child anxiety that views mother and child attachment beliefs and children's perceptions of maternal control as mediators of the association. The study was conducted with mothers and their children aged 6 to 17 (N = 88). Maternal anxiety was significantly associated with child anxiety and maternal anxious attachment beliefs, whereas child anxiety was associated with maternal anxious attachment beliefs, child insecure attachment beliefs, and children's perceptions of maternal control. Maternal anxious attachment beliefs mediated the association between maternal and child anxiety; however, child attachment beliefs and children's perceptions of maternal control did not. The influence of age and gender is also addressed. The findings are discussed in terms of elucidating the role of attachment beliefs and children's perceptions of maternal control on the association between maternal and child anxiety.     

Application of classification trees to the habitat preference of upland birds

This paper investigates the application of classification tree methodology to the bird habitat associations of the red grouse in the south Pennines of England. The effectiveness of this classification will be investigated using the kappa coefficient. Although the results for these data are disappointing, the methodology seems to have potential for this kind of problem.     

Ultrastructural elliptical models

Dolby's (1976) ultrastructural model with no replications is investigated within the class of the elliptical distributions. General asymptotic results are given for the sample covariance matrix S in the presence of incidental parameters. These results are used to study the asymptotic behaviour of some estimators of the slope parameter, unifying and extending existing results in the literature. In particular, under some regularity conditions they are shown to be consistent and asymptotically normal. For the special case of the structural model, some asymptotic relative efficiencies are also reported which show that generalized least squares and the method of moment estimators can be highly inefficient under nonnormality.     

Determination of a Tolerable Daily Intake of DDT for Consumers of DDT Contaminated Fish from the Lower Yakima River, Washington*

DDT, DDE, and DDD have been detected at elevated concentrations in sediments and fish of the Yakima River, its tributaries and drainages. An assessment was conducted to evaluate the public health significance of eating fish from the river. This was accomplished by establishing a daily intake level of DDT for the population of greatest concern, and comparing this level to a tolerable daily intake. The most sensitive and highly exposed group was determined to be breastfeeding infants. Infant daily intakes of DDT, based on estimated mother's DDT-breast milk levels, were compared to a recommended tolerable daily intake. Results indicate that mothers who frequently consume Yakima River bottom-feeding fish could have breast milk DDT concentrations sufficiently high to expose their infants to levels above the tolerable daily intake.     

Bayesian prediction for progressively censored data from the Burr model

Received: August 14, 2000; revised version: April 23, 2001     

The Obese Smoker's Wage Penalty*

Objective. Smoking and obesity are associated with significant wage penalties when considered separately. We assess the combined effects of those behaviors. Methods. We estimate the effects of smoking and obesity on wages using multivariate regression analysis with 1979 National Longitudinal Survey of Youth data. Results. The raw data indicate that obese smokers experience large earnings penalties. However, these penalties are not found to be interactive or directly causal for most of the demographic subgroups we examine. One exception is the wage penalty associated with obesity for females, which remains significant throughout the analysis. Conclusions. In the absence of a demonstrable direct causal effect of those behaviors on wages, associated individual‐specific socioeconomic factors appear to be the driving forces behind the obese smoker wage penalty. Not included, but potentially significant, are the effects of employer and customer discrimination against obese smokers and the possible labor policy implications of such discrimination.     

A Markov Chain Monte Carlo version of the genetic algorithm Differential Evolution: easy Bayesian computing for real parameter spaces

Differential Evolution (DE) is a simple genetic algorithm for numerical optimization in real parameter spaces. In a statistical context one would not just want the optimum but also its uncertainty. The uncertainty distribution can be obtained by a Bayesian analysis (after specifying prior and likelihood) using Markov Chain Monte Carlo (MCMC) simulation. This paper integrates the essential ideas of DE and MCMC, resulting in Differential Evolution Markov Chain (DE-MC). DE-MC is a population MCMC algorithm, in which multiple chains are run in parallel. DE-MC solves an important problem in MCMC, namely that of choosing an appropriate scale and orientation for the jumping distribution. In DE-MC the jumps are simply a fixed multiple of the differences of two random parameter vectors that are currently in the population. The selection process of DE-MC works via the usual Metropolis ratio which defines the probability with which a proposal is accepted. In tests with known uncertainty distributions, the efficiency of DE-MC with respect to random walk Metropolis with optimal multivariate Normal jumps ranged from 68% for small population sizes to 100% for large population sizes and even to 500% for the 97.5% point of a variable from a 50-dimensional Student distribution. Two Bayesian examples illustrate the potential of DE-MC in practice. DE-MC is shown to facilitate multidimensional updates in a multi-chain “Metropolis-within-Gibbs” sampling approach. The advantage of DE-MC over conventional MCMC are simplicity, speed of calculation and convergence, even for nearly collinear parameters and multimodal densities.     

A New Look at Half-normal Plots for Assessing the Significance of Contrasts for Unreplicated Factorials

The visual inspection of a half-normal plot is a popular procedure for interpreting data from unreplicated factorial experiments. As this technique is somewhat subjective, the plot should preferably be supplemented with a more formal procedure. However, despite the improvements proposed by Zahn, the formal method using the half-normal plot suggested by Daniel seems to be rarely used. We investigate this procedure by using simulations and real data and suggest a modification to improve its acceptability. We also present corrected critical values for a procedure proposed by Lenth which was found to be miscalibrated.