Robust nonparametric tests for the two-sample location problem

We construct and investigate robust nonparametric tests for the two-sample location problem. A test based on a suitable scaling of the median of the set of differences between the two samples, which is the Hodges-Lehmann shift estimator corresponding to the Wilcoxon two-sample rank test, leads to higher robustness against outliers than the Wilcoxon test itself, while preserving its efficiency under a broad range of distributions. The good performance of the constructed test is investigated under different distributions and outlier configurations and compared to alternatives like the two-sample t-, the Wilcoxon and the median test, as well as to tests based on the difference of the sample medians or the one-sample Hodges-Lehmann estimators.     

Comparison of multiple imputation and two-phase logistic regression to analyse two-phase case–control studies with rich phase 1: a simulation study

Two-phase case–control studies cope with the problem of confounding by obtaining required additional information for a subset (phase 2) of all individuals (phase 1). Nowadays, studies with rich phase 1 data are available where only few unmeasured confounders need to be obtained in phase 2. The extended conditional maximum likelihood (ECML) approach in two-phase logistic regression is a novel method to analyse such data. Alternatively, two-phase case–control studies can be analysed by multiple imputation (MI), where phase 2 information for individuals included in phase 1 is treated as missing. We conducted a simulation of two-phase studies, where we compared the performance of ECML and MI in typical scenarios with rich phase 1. Regarding exposure effect, MI was less biased and more precise than ECML. Furthermore, ECML was sensitive against misspecification of the participation model. We therefore recommend MI to analyse two-phase case–control studies in situations with rich phase 1 data.     

Demokratie für den Irak?

The situation in Iraq remains unstable. This is, among other factors, due to the lack of a realistic, long-term and integrative model for democratization. The current occupying powers neither dispose of a long-term plan, appropriate to the local circumstances, which could be able to include systematically local culture and social forms. Nor do they obviously have a clear step-by-step plan within the framework of an integral model which could take into account the experience of the past decades regarding sustainable democratization of Islamic core areas. This essay presents the three currently most promising and most realistic models of how Iraq can be democratized by means of a long-term, comprehensive, holistic and systematically differentiated strategy. The focus of attention is on the methodical inclusion of cultural, socio-psychological and sociological aspects. The three democratization models for Iraq are: 1) the Tatarstan model, 2) the Turkey model, 3) the model of stratified democracy. These three models are critically discussed and evaluated; in conclusion a short outlook will be drafted.     

Bayesian design in queues: An application to aeronautic maintenance

We exploit Bayesian criteria for designing M/M/c//r$M/M/c//r$ queueing systems with spares. For illustration of our approach we use a real problem from aeronautic maintenance, where the numbers of repair crews and spare planes must be sufficiently large to meet the necessary operational capacity. Bayesian guarantees for this to happen can be given using predictive or posterior distributions.     

Bilanzmanipulation und Managementpräferenzen bezüglich der Strenge des Systems der Rechnungslegungsregulierung

The present paper analyses the political economy of accounting regulation. It develops a game theoretic model of earnings management in a capital market context. Various forms of information asymmetry are examined. The analysis shows that the preferences of managers over the strictness of accounting regulation depend on the nature of the information asymmetry. In a number of instances, managers prefer to be subject to strict regulation. This result contradicts the often heard assertion that managers prefer lax regulation and has implications for regulatory competition.     

Risk Management in a Polio-Free World

Inherent in the decision to launch the Global Polio Eradication Initiative in 1988 was the expectation for many people that immunization against poliomyelitis would eventually simply stop, as had been the case with smallpox following its eradication in 1977. However, the strategies for managing the risks associated with a "polio-free" world must be continuously refined to reflect new developments, particularly in our understanding of the live polioviruses in the oral poliovirus vaccine (OPV) and in the international approach to managing potential biohazards. The most important of these developments has been the confirmation in 2000 that vaccine-derived polioviruses (VDPVs) can circulate and cause polio outbreaks, making the use of OPV after interruption of wild poliovirus transmission incompatible with a polio-free world. A comprehensive strategy has been developed to minimize the risks associated with eventual OPV cessation, centered on appropriate long-term biocontainment of poliovirus stocks (whether for vaccine production, diagnosis, or research), the controlled reintroduction of any live poliovirus vaccine (i.e., from an OPV stockpile), and appropriate use of the inactivated poliovirus vaccine (IPV). Although some aspects of this risk management strategy are still debated, there is wide agreement that no strategy would entirely eliminate the potential risks to a polio-free world. The current strategy for risk management in a polio-free world will continue to evolve with better characterization of these risks and the development of more effective approaches both to reduce those risks and to limit their consequences should they occur.     

Health Risk Assessment After Decontamination of the Beaches Polluted by the Wrecked ERIKA Tanker

Following the wreck of the oil tanker ERIKA off the north-west coast of France in December 1999, cleaning up of the beaches involved considerable work, which in any case could not be perfect. This raised the question of the short- and long-term health risks for the future bathers related to the toxicity of the remaining oil polycyclic aromatic hydrocarbons (PAHs). This risk assessment study was conducted to help health authorities plan risk management policies and inform the public. Thirty-six beaches were selected, representing a sample of the most frequently encountered topographic and beach usage situations; seven "control" beaches, unspoiled by ERIKA, were also investigated. Samples of water and sand were taken from each site, as well as from the surface of rocks. The 16 PAHs selected by the U.S. EPA were quantified in each environment. Several scenarios of exposure were contemplated: (1) a child between 2 and 4 years accidentally ingesting a small ball of fuel, (2) a child daily exposed throughout his holiday-time stay, (3) an adult (including a pregnant woman) spending his/her holidays on the coast, (4) an adult working on the beach, (5) and an adult practicing water sports. Among the available and significant toxicological values, the most conservative ones were selected for computing risks. The sand and water, after decontamination, were slightly polluted (respectively, 7.8 microg/kg and 23.3 ng/l of total 16 PAHs), with values similar to those found in the control beaches. By contrast, the rocky areas in some places were still highly polluted (up to 23 mg/kg on the surface layer). No lethal risk was found for a young child who had accidentally ingested a small ball of fuel. The life-long excess risks for skin cancer and for all other cancers were about 10-5 in scenarios including contact with the polluted rocks. In all other cases, excess risks were considerably lower. The hazard quotient for teratogenic effects was very small, except in scenarios where pregnant women would walk among rocks containing high pollution levels. In conclusion, exposure was mainly associated with polluted water among children, and with spoiled rocks for adults. Despite uncertainties, mainly dealing with the prediction of long-term risks following a short-term exposure, this study showed that beaches where pollution was no longer visible after decontamination did not entail any significant health risks and could be opened to the public.     

Gambling behavior and psychological disorders of gamblers on German-style slot-machines

As part of a longitudinal study on gambling behavior, 300 gamblers who differed in the extent of their subjective stress were surveyed by clinical psychologists. Self ratings, interviewer ratings and questionnaires were used to assess socio-demographics, gambling behavior, psychological disturbance and social consequences. Our results lead to two central conclusions: 1. The amount of time spent on gambling activities per week and the number of years since starting gambling were not significant predictors of the extent of subjective stress related to gambling. 2. There are various sources for the subjective stress of gambling, such as marital discord and lack of social assertiveness. In a subgroup, stress was directly related to gambling. Generally, we found a higher incidence of alcohol abuse and depression in the stressed gamblers. Further research is indicated to obtain information on the etiology and development of these disturbances.This article was first published in Germany: Törne v. I, Konstanty, R: (1989), Spielverhalten und Störungsbilder bei Spielern an Geldspielautomaten, Suchtgefahren, 35, 14–34.     

Rethinking population distribution policies

Developed and developing nations are generally dissatisfied with the spatial distributions of their populations. Dissatisfaction is particularly acute on the part of developing nations which view their problems of population distribution as more serious than those of natural increase. The overwhelming majority of governments have already adopted policies to affect rates and patterns of internal migration or the configuration of their rural and urban populations. However, recent reviews of these policies suggest they have been only partially successful. Apparent policy deficiencies are reviewed and possible remedies suggested in regard to the need for intervention, the scope of policies, their objectives, instruments, and evaluation.