A family of distributions to model load sharing systems

The main characteristic of a load sharing system is that after the failure of one component the surviving components have to shoulder extra load and hence are prone to failure at an earlier time than what is expected under the original model. In others, the failure of one component may release extra resources to the survivors, thus delaying the system failure. In this paper we consider such m component systems and some observation schemes and identifiability issues under them. Then we construct a general semiparametric multivariate family of distributions which explicitly models this phenomenon through proportional conditional hazards. We suggest estimates for the constant of proportionality. We propose a nonparametric test for the hypothesis that the failures take place independently according to the common distribution against the alternative hypothesis that the second failure takes place earlier than warranted, study its properties and illustrate its use.     

Modeling statistical dependence of Markov chains via copula models

Conditional probability distributions have been commonly used in modeling Markov chains. In this paper we consider an alternative approach based on copulas to investigate Markov-type dependence structures. Based on the realization of a single Markov chain, we estimate the parameters using one- and two-stage estimation procedures. We derive asymptotic properties of the marginal and copula parameter estimators and compare performance of the estimation procedures based on Monte Carlo simulations. At low and moderate dependence structures the two-stage estimation has comparable performance as the maximum likelihood estimation. In addition we propose a parametric pseudo-likelihood ratio test for copula model selection under the two-stage procedure. We apply the proposed methods to an environmental data set.     

Themes from older men's lay definitions of successful aging as indicators of primary and secondary control beliefs over time: The Manitoba Follow-up Study

Constructs of control have theoretically been equated to successful aging in the psychology literature. Hence, we used themes from lay definitions of successful aging to quantify the prevalence of primary and secondary control beliefs over time. In doing so we hoped to shed new light upon the virtually uncharted area of older men's primary and secondary control beliefs over time. Using successful aging narratives spanning a 10-year timeframe from the Manitoba Follow-up Study cohort, we mapped themes from older men's lay definitions of successful aging onto Rothbaum, Weisz, and Snyder's (1982) constructs of primary and secondary control. We then examined the prevalence of the constructs of control over 10 years and found that some men emphasized primary control, some emphasized secondary control, and others emphasized both, prospectively. Counter to what had previously been theorized, many older men continued to emphasize primary control as important well into late life. As expected, secondary control became more important with age. Furthermore, among those men who endorsed both primary and secondary control, significantly more men switched emphasis from primary to secondary control beliefs as they aged. This finding supported Rothbaum et al.'s (1982) surmise that individuals could switch from one type of control to another, presumably as life circumstances dictated. Knowing which types of control beliefs older men emphasize as they age has theoretical and practical implications. Theoretically, it sheds new light on the under-researched area of control beliefs in older men. Practically, it is informative for anyone interested in enhancing older men's perceptions of control in very late life, particularly in the face of otherwise uncontrollable age-related decline and imminent demise.     

Macroeconomic determinants of migration: the case of Germany, 1964-1988

Macroeconomic determinants of immigration are analysed with pooled cross-country and time series data on net immigration to Germany from European countries during 1964-1988.
Results demonstrate clearly that both high and low frequency determinants have been important. Long run trends are determined by the degree and speed of per capita income convergence between the sending and receiving countries, while year-to-year changes in net immigration flows are dominated by cyclical economic conditions.     

Health care trends, Part 2. The new health care team. Panel discussion

In Part 2 of this third annual panel discussion, six experts talk about the growing diversity of health care providers and what it means for consumers and physicians. Americans are getting their wellness and health care services from a wider variety of non-physician practitioners than ever before. The number of allied health and alternative providers with direct patient access is likely to continue growing. This trend is being driven by consumer demand, by the lobbying efforts of non-physician providers, and by federal, state, and private payers who see the potential for reduced health care spending, greater consumer satisfaction, and better outcomes. In practice, this means physicians and non-physician providers, some of whom may not be sanctioned by the medical establishment, are obligated to collaborate as a team. Members of this new provider team will have to communicate effectively (with each other, with consumers, and with payers) and make evidence-based clinical decisions. Physicians may have to share decision-making with other members of this new health care team.     

Consumer Evaluations of Food Risk Management Quality in Europe

In developing and implementing appropriate food risk management strategies, it is important to understand how consumers evaluate the quality of food risk management practices. The aim of this study is to model the underlying psychological factors influencing consumer evaluations of food risk management quality using structural equation modeling techniques (SEM), and to examine the extent to which the influence of these factors is country-specific (comparing respondents from Denmark, Germany, Greece, Slovenia, and the United Kingdom). A survey was developed to model the factors that drive consumer evaluations of food risk management practices and their relative importance (n= 2,533 total respondents). The measurement scales included in the structural model were configurally and metrically invariant across countries. Results show that some factors appear to drive perceptions of effective food risk management in all the countries studied, such as proactive consumer protection, which was positively related to consumers' evaluation of food risk management quality, while opaque and reactive risk management was negatively related to perceived food risk management quality. Other factors appeared to apply only in certain countries. For example, skepticism in risk assessment and communication practices was negatively related to food risk management quality, particularly so in the UK. Expertise of food risk managers appeared to be a key factor in consumers' evaluation of food risk management quality in some countries. However, trust in the honesty of food risk managers did not have a significant effect on food risk management quality. From the results, policy implications for food risk management are discussed and important directions for future research are identified.     

Effect of Level of Disaggregation on Conjoint Cross Validations: Some Comparative Findings*

Early formulations of conjoint models focused on part-worth estimation at the individual level. As the methodology's popularity grew so did industry demands for increasingly larger numbers of attributes and levels. In response to these demands, new approaches, based on partial or full data aggregation (such as clusterwise/latent class conjoint and choice-based conjoint), have appeared. This paper suggests that pooled-data models will often be successful in predicting market shares when researchers employ monotonic attributes. In these cases more of a good attribute (or less of a bad attribute) is always more preferred. In the more realistic case, in which some of the attributes may be nonmonotonic, we find that data aggregation does not predict holdout sample preferences as well as individual part-worth models.