Bayesian model learning based on a parallel MCMC strategy

We introduce a novel Markov chain Monte Carlo algorithm for estimation of posterior probabilities over discrete model spaces. Our learning approach is applicable to families of models for which the marginal likelihood can be analytically calculated, either exactly or approximately, given any fixed structure. It is argued that for certain model neighborhood structures, the ordinary reversible Metropolis-Hastings algorithm does not yield an appropriate solution to the estimation problem. Therefore, we develop an alternative, non-reversible algorithm which can avoid the scaling effect of the neighborhood. To efficiently explore a model space, a finite number of interacting parallel stochastic processes is utilized. Our interaction scheme enables exploration of several local neighborhoods of a model space simultaneously, while it prevents the absorption of any particular process to a relatively inferior state. We illustrate the advantages of our method by an application to a classification model. In particular, we use an extensive bacterial database and compare our results with results obtained by different methods for the same data.     

Questioning the Notion of Feminine Leadership: A Critical Perspective on the Gender Labelling of Leadership

Traditionally, leadership has been equated with masculinity. Managerial jobs, at least in business and on senior levels, have been defined as a matter of instrumentality, autonomy, result‐orientation, etc. something which is not particularly much in line with what is broadly assumed to be typical for females. Today, however, there seems to be a broad interest in leadership being more participatory, non‐hierarchical, flexible and group‐oriented. These new ideas on leadership are often seen by students of gender as indicating a feminine orientation. This article argues that it is necessary to critically discuss the whole idea of gender labelling leadership as masculine or feminine and suggests that we should be very careful and potentially aware of the unfortunate consequences when we use gender labels. Constructing leadership as feminine may be of some value as a contrast to conventional ideas on leadership and management but may also create a misleading impression of women's orientation to leadership as well as reproducing stereotypes and the traditional gender division of labour.     

Alcohol-Related Mortality in 15 European Countries in the Postwar Period

The aim of this paper isto assess postwar differences and trends inalcohol-related mortality in the currentEuropean Union (minus Luxembourg plus Norway)on the basis of liver cirrhosis mortality anddeaths with explicit mention of alcohol,primarily alcohol dependence, alcohol psychosisand alcohol poisoning (AAA). The questionof the extent to which these indicators arecomparable across Western European countries isalso addressed. A marked north-south gradientwas found for cirrhosis mortality, with thehighest rates revealed in Southern Europe andthe lowest in Northern Europe. However, thisgradient weakened with the passage of time andthe initially quite substantial regionaldifferences declined during the latter part ofthe study period. Explicitly alcohol-relatedmortality (AAA), on the other hand, showed areverse cross-national pattern with the highestrates in the north and the lowest in the south.A positive cross-national relationship wasobserved between cirrhosis and per capitaconsumption but this match was not improved bycombining cirrhosis with explicitlyalcohol-related causes. Nevertheless, withinSouthern, Central and Northern Europeancountries the relationship between per capitaconsumption and AAA-mortality was positive. Itis concluded that cirrhosis mortality is usefulfor making rough national comparisons in aWestern European context whereas the validityof explicitly alcohol-related mortality isquestionable. Cultural differences in recordingpractices and drinking patterns are discussedas possible determinants of geographicaldifferences in AAA-mortality.     

Expected and unexpected values of individual Mahalanobis distances

This paper derives first-order sampling moments of individual Mahalanobis distances (MDs) in cases when the dimension p of the variable is proportional to the sample size n. Asymptotic expected values when n, p → ∞ are derived under the assumption p/n → c,?0 ? c < 1. It is shown that some types of standard estimators remain unbiased in this case, while others are asymptotically biased, a property that appears to be unnoticed in the literature. Second-order moments are also supplied to give some additional insight to the matter.     

Estimating mean-standard deviation ratios of financial data

This article treats the problem of linking the relation between excess return and risk of financial assets when the returns follow a factor structure. The authors propose three different estimators and their consistencies are established in cases when the number of assets in the cross-section (n) and the number of observations over time (T) are of comparable size. An empirical investigation is conducted on the Stockholm stock exchange market where the mean-standard deviation ratio is calculated for small- mid- and large cap segments, respectively.     

Taxes and Quantity Constraints in a Model of Male Labour Supply in Sweden

The purpose of this study is to estimate a model of male labour supply, when agents may be subject to quantity constraints. As the data contain direct information about hours constraints, this information is used in the estimation. We also quantify the importance of such rationing for the response in the hours of work to changes in wages, income and the tax system. As it turns out, quantity constraints appear only to slightly reduce the sensitivity of the average hours of work to changes in the economic environment.     

Multiattribute Risk Analysis in Nuclear Emergency Management

Radiation protection authorities have seen a potential for applying multiattribute risk analysis in nuclear emergency management and planning to deal with conflicting objectives, different parties involved, and uncertainties. This type of approach is expected to help in the following areas: to ensure that all relevant attributes are considered in decision making; to enhance communication between the concerned parties, including the public; and to provide a method for explicitly including risk analysis in the process. A multiattribute utility theory analysis was used to select a strategy for protecting the population after a simulated nuclear accident. The value-focused approach and the use of a neutral facilitator were identified as being useful.