Minimum and Maximum Information Censoring Plans in Progressive Censoring

Expressions for the entropy, the Kullback-Leibler information, and the I _α-information are established for distributions of progressively Type-II censored order statistics. These results are used to identify minimum and maximum information censoring plans. In particular, we find minimum and maximum entropy plans for DFR, exponential, Pareto, reflected power, and Weibull distributions. The results for Kullback-Leibler and I _α-information hold for any continuous distribution.     

Peer-led and adult-led programs--student perceptions

This research examines whether peer-led prevention programs are preferable to adult-led programs. Participants were 2,447 students in 94 classes, from 31 schools running drug prevention programs. The schools were divided into two groups according to the model they used in their program: fifteen schools used peer-led model, while sixteen used the adult-led model. A 46-item questionnaire was constructed in order to examine the students' perception of the programs. The results show that all the input measures (content, atmosphere, openness, discipline, facilitators' competence) and the outcome measures (satisfaction, knowledge, avoidance, curiosity, personal relationship) were perceived as more positive in the peer-led model. The differences were small, but significant. While the findings suggest that the peer-led model has a somewhat greater potential for primary prevention, the differences found do not enable us to state with certainty that this model is preferable for primary prevention purposes.     

Extreme Value Analysis for Progressively Type-II Censored Order Statistics

An account to extreme value theory for progressively Type-II censored order statistics is presented which enables us to handle limit laws for upper and lower extreme, intermediate and central progressively Type-II censored order statistics within one framework. We illustrate that the extreme value analysis for progressively Type-II censored order statistics is connected to limit laws for sums of independent but not-identically distributed exponential random variables. Moreover, we show that the limits are transformations of extreme value distributions and illustrate the connection to extreme value analysis for order statistics.     

Behaviour of the length test for medium sample sizes

In this note it is shown that even for relatively large sample sites the asymptotic distribution of the smoothed length as derived in Reschenhofer and Bomse (1991) should not be used for the determination of critical values. Therefore extended tables of critical values for both the 1% and 5% levels of significance generated by simulation are presented.     

Europe needs a better crisis management system

I proposes a comprehensive approach towards crisis management and explore the relation of crisis management to European politics. Drawing insight from my work as Special Co-ordinator of the Stability Pact for South Eastern Europe, I propose ten lessons for crisis management that could provide a perspective for both the future of conflict prevention and the European project. The profound changes that have affected Europe over the last decade have put several assumptions about the nature of crisis management in question and have asked for a clear perspective for the European project. I will try to elaborate on an integrative approach towards crisis management including a broadened concept of security and its relation to the economy and the public sphere. The current crisis within the European Community concerning its identity and a perspective for the future is used as an incentive for defining a position and challenging the problems that lie ahead.