MVE, MCD, and MZE: A simulation study comparing convex body minimizers

Summary: In a recent work (Paris Scholz, 2002), a new robust estimator for convex bodies has been proposed, based on the estimation of a zonoid of a distribution. This so–called minimum volume zonoid estimator (MZE) is similar in type to the well–known robust approaches of the minimum volume ellipsoid (MVE) and the minimum covariance determinant (MCD), all three seeking for a subset of given data for which some criteria are minimized. We investigate the similarity between these three concepts by comparing which subsets are chosen to be the optimal ones.     

Lessons from history: pasts,presents and futures of punishment in children's talk

This paper draws on a study of the ways in which the moral and practical dilemmas of punishment are debated and deliberated upon in discussions among nine year old children (with adult facilitators). Theoretically, we are concerned with the points of connection between the social study of childhood and the analysis of punishment as an arena of discourse and practice; and methodologically we address this topic from the perspective of conversation. Here we focus on one aspect of the children's talk, namely the mobilisation of themes and images that seem to have their origin in the more or less remote past. We argue that a contextual and nuanced exploration of the historicity of such ‘punishment talk’ can help in disclosing how punishment ‘works’ as a theme in politics and culture—history offers a resource of precedents and examples that children find good to think with. We suggest a particular association between this historical imagination (and the often physically forceful measures that it invokes) and times of crisis or emergency. This is apparent, for example, when discussion turns to the persistence of the death penalty in the United States. We conclude with some brief remarks on the significance of involving children in such deliberations for the development of alternative (or ‘replacement’) penal politics. Copyright © 2002 John Wiley & Sons, Ltd.     

Towards a unified definition of maximum likelihood

A unified definition of maximum likelihood (ml) is given. It is based on a pairwise comparison of probability measures near the observed data point. This definition does not suffer from the usual inadequacies of earlier definitions, i.e., it does not depend on the choice of a density version in the dominated case. The definition covers the undominated case as well, i.e., it provides a consistent approach to nonparametric ml problems, which heretofore have been solved on a more less ad hoc basis. It is shown that the new ml definition is a true extension of the classical ml approach, as it is practiced in the dominated case. Hence the classical methodology can simply be subsumed. Parametric and nonparametric examples are discussed.     

Regular Adjustment of Financial Parameters of Social Protection Systems in Volatile Inflationary Environments

This paper addresses the problem of periodical adjustments of financial parameters which govern the revenue side of social protection systems and adjustments of social benefits-in-payment in a volatile visionary environment: in short, the indexation problem. It attempts to provide an insight into this problem by focusing on cyclical wage movements, usually ignored in economic and actuarial projections. Preliminary results are drawn from two numerical examples, showing impacts on social security finances, equity and consumption. Some general conclusions are drawn, notably the problem affects both pay-as-you-go and funded systems. Essentially, in terms of adjustment as in other spheres, when it comes to addressing problems in the present, past experience is not necessarily the best basis for constructing a rational future.     

Risk Perception of Heavy Metal Soil Contamination by High-Exposed and Low-Exposed Inhabitants: The Role of Knowledge and Emotional Concerns

Soil contaminated with heavy metals is a salient example of environmental risk. Consumption of vegetables cultivated in contaminated soil or direct ingestion of soil by small children can damage health. In contrast to other kinds of pollution or risks such as air pollution or exposure to ozone, the individual risk concerning soil contamination is highly dependent on the way one is exposed to the local source of risk. Thus, we wanted to know if risk perception varies according to the level of exposure. A quasi-experimental, questionnaire-based study was conducted in a community in northwest Switzerland, where the soil is widely contaminated. The level of contamination varies with the distance from the source of the contamination, a metal processing plant. We investigated the perception of risk of heavy-metal-contaminated soil by inhabitants with high-exposure levels (N= 27) and those with low-exposure levels (N= 30). Both groups judged the risk for oneself similarly whereas the low-exposure group, when compared to the high-exposure group, judged perceived risk for other affected people living in their community to be higher. Besides this exposure effect, risk perception was mainly determined by emotional concerns. Participants with higher scores in self-estimated knowledge tended to provide low-risk judgments, were less interested in further information, showed low emotional concern, and thus displayed high risk acceptance. In contrast, actual knowledge showed no correlation with any of theses variables. Judgments on the need for decontamination are determined by risk perception, less application of dissonance-reducing heuristics and commitment to sustainability. The desire for additional information is not affected by missing knowledge but is affected by emotional concerns.     

Risk Perception of Heavy Metal Soil Contamination and Attitudes toward Decontamination Strategies

Contaminated soils are a common environmental risk all over the world. One major source of risk is heavy metal soil contamination caused by industrial emissions. This quasiexperimental study investigated the perception of these risks by exposed and nonexposed people, their attitudes toward bioremediation methods using hyperaccumulating plants, and the influence of long-term aspects of sustainability on the acceptance of bioremediation methods. Major findings were that people living in a contaminated area perceived the risk of the heavy metal soil contamination as higher than the general risk of contamination. Second, a factor analysis showed that the factors dread, control, and catastrophic potential were relevant for the perception and valuation of low-dose environmental risks such as the contamination of the investigated area. In addition, a cluster analysis showed that the risk of heavy metal soil contamination was perceived as similar to that of oil contamination, ozone layer, preservatives and genetic technology. It was perceived indifferently with regard to dread. The uncontrollability of heavy metal soil contamination was estimated as medium, and its catastrophic potential as low. Third, exposed and nonexposed participants preferred bioremediation methods to classical methods (e.g., excavation and chemical treatment of the soil), because they perceived the environmental and esthetical performance of the bioremediation as important criteria. Sustainability or precautionary issues, such as the prevention of harm for future generations, were highly correlated with the acceptance of the use of bioremediation methods in people's residential areas.     

Measuring societal effects of transdisciplinary research projects: design and application of an evaluation method

Most Transdisciplinary Research (TdR) projects combine scientific research with the building of decision making capacity for the involved stakeholders. These projects usually deal with complex, societally relevant, real-world problems. This paper focuses on TdR projects, which integrate the knowledge of researchers and stakeholders in a collaborative transdisciplinary process through structured methods of mutual learning. Previous research on the evaluation of TdR has insufficiently explored the intended effects of transdisciplinary processes on the real world (societal effects). We developed an evaluation framework for assessing the societal effects of transdisciplinary processes. Outputs (measured as procedural and product-related involvement of the stakeholders), impacts (intermediate effects connecting outputs and outcomes) and outcomes (enhanced decision making capacity) are distinguished as three types of societal effects. Our model links outputs and outcomes of transdisciplinary processes via the impacts using a mediating variables approach. We applied this model in an ex post evaluation of a transdisciplinary process. 84 out of 188 agents participated in a survey. The results show significant mediation effects of the two impacts "network building" and "transformation knowledge". These results indicate an influence of a transdisciplinary process on the decision making capacity of stakeholders, especially through social network building and the generation of knowledge relevant for action.