On the distribution of the test statistics of Kendall and Wilcoxon's two sample test when ties are present

Zoals bekend is, kan de exacte verdeling van de toetsingsgrootheid van Kendall voor rangcorrelatie als er geen gelijke waarnemingen optreden en ah de nulhypothese juist is, berekend worden met behulp van een recurrente formule. Hetzelfde geldt voor de toetsingsgrootheid van Wilcoxon voor het geval van twee steekproeven. Wanneer er wél gelijke waarnemingen optreden, kan men, zoals hier wordt aangetoond, een gegeneraliseerde recursieformule gebruiken. De berekeningsmethode, die berust op het herhaald toepassen van deze recursieformule, is minder afschrikwekkend dan die welke bestaat in het opstellen van alle mogelijke schema's en de berekening van de bijbehorende kansen.     

People’s responses to risks of electromagnetic fields and trust in government policy: the role of perceived risk,benefits and control

Trust in government policy affects the way people perceive and handle risks. In our study, we investigated the relationships between trust in government policy regarding electromagnetic fields (EMF), perceived risk and perceived benefits of public and personal EMF sources, perceived control over exposure to EMF and responses to the possible EMF health risk (e.g. protest against placement of mobile phone base stations or power lines, or taking own measures against EMF exposure). Previous research indicated that perceived risk and benefits mediate the relationship between trust and people’s risk responses. Additionally, we suggest that perceived control over EMF exposure affects the relation between trust in government policy and perceived risk, and, consequently, the risk responses. We performed a survey among the Dutch population (n?=?1009), which contained questions about risk responses to EMF, perceived risk and benefits of several EMF sources, trust in government policy and perceived control over EMF exposure. Comparing public EMF sources, i.e. power lines and mobile phone base stations, to personal EMF sources, i.e. microwave ovens and cordless and mobile phones, we tested our hypotheses. Variations in risk responses to both public and personal EMF sources were mainly explained by risk perception. In addition, perceived risk partially mediated the relationship between trust in government policy and risk responses. For public sources, perceived control over exposure weakened the negative relationship between trust and perceived risk. We conclude that, especially in people with low perceived control, a lack of trust in government policy may enhance perceptions of health risks, thereby increasing their inclination for risk responses.     

Horizontal integration in the Dutch financial sector

In this paper, the consequences of cross-shareholding in an n-firm industry are analyzed. Our attention focuses on the case where firms have silent interests in each other. These interests can be direct or indirect. We analyze the effects of cross-shareholding on the price–cost margins in a Cournot and a Bertrand setting. In all cases, competition is reduced due to shareholding interlocks. As an empirical example the Dutch financial sector is used. Comparing the case of shareholding with the case of no-shareholding, the price–cost margins are found to be up to 2% higher in a Bertrand market, and at least 8% higher in a Cournot market.