Analyses of multiple types of online gambling within one provider: an extended evaluation framework of actual online gambling behaviour

The aim of this article is to extend previous publications of actual online gambling behaviour that neglected involvement across multiple types of gambling and did not provide levels of at-risk involvement. Behavioural data from 27,653 subscribers of an online gambling provider (bwin) in February 2005 were reanalysed across eight products over seven months. Established involvement levels of offline gambling segregated possible online at-risk gamblers. Forty-seven percent of the sample exceeded at least one of the two most conservative thresholds. Each additionally used gambling product increased the risk of transgressing involvement cut-offs by 3.06 to 4.27 times, but type-specific risks decreased strongly after adjusting for involvement in multiple gambling types. Only Poker and Live-action betting remained significant risk factors after adjustment. Taken together, cross-product analyses of gambling patterns lay the groundwork for an extended understanding of individual online gambling behaviour and overcome the methodological artefacts of isolated analyses.     

Does Religion Make You Healthier and Longer Lived? Evidence for Germany

Researchers in the US have consistently reported substantial—not just statistically significant—links between religious belief and practice, and improved health and longevity. In this paper we report evidence for Germany, using data from the long-running, nationally representative German Socio-Economic Panel (SOEP 1984). The SOEP dataset includes multiple measures of health, plus many ‘controls’ which it is appropriate to use in assessing links between religious practice, health and longevity. These controls include personality traits known to be associated with better health (notably conscientiousness), and also the age of death of parents of the survey respondents. Initial results suggested that religious practice (church attendance) may be linked only to subjective (self-rated) measures of health, not to more objective measures. It seemed possible that results in some previous research could be due to what may be termed satisfaction bias or positivity bias; the known tendency of religious people to report higher than average satisfaction with almost all aspects of life. Further investigation indicated that relationships between church attendance and subjective measures of health were weaker, when a control for satisfaction bias was in place. However, there was countervailing evidence that the subjective measures in SOEP may actually be more not less valid than the objective measures; they are better not worse predictors of mortality. It was also clear that religious belief and church attendance are associated with health-protective behaviors and attitudes, including taking more exercise, not smoking and higher life satisfaction. At the end of the paper we estimate a structural equation model which maps links between religious practice, these protective behaviors and attitudes, and improved health outcomes.     

Weak representation of the cumulative hazard function under semiparametric random censorship models

In this paper we derive a weak representation of the semiparametric estimator A^se _nof the cumulative hazard function A in the random censorship model. Based on this representation we show that |A^se _n- A| is uniformly bounded in probability up to the last order statistic of the observations.     

On cross-validation for discrete kernel estimates in discrimination

The choice of smoothing determines the properties of nonparametric estimates of probability densities. In the discrimination problem, the choice is often tied to loss functions. A framework for the cross–validatory choice of smoothing parameters based on general loss functions is given. Several loss functions are considered as special cases. In particular, a family of loss functions, which is connected to discrimination problems, is directly related to measures of performance used in discrimination. Consistency results are given for a general class of loss functions which comprise this family of discriminant loss functions.     

A characterization of the single-crossing domain

We characterize single-crossing preference profiles in terms of two forbidden substructures, one of which contains three voters and six (not necessarily distinct) alternatives, and one of which contains four voters and four (not necessarily distinct) alternatives. We also provide an efficient way to decide whether a preference profile is single-crossing.     

Inverse censoring weighted median regression

We implement semiparametric random censorship model aided inference for censored median regression models. This is based on the idea that, when the censoring is specified by a common distribution, a semiparametric survival function estimator acts as an improved weight in the so-called inverse censoring weighted estimating function. We show that the proposed method will always produce estimates of the model parameters that are as good as or better than an existing estimator based on the traditional Kaplan–Meier weights. We also provide an illustration of the method through an analysis of a lung cancer data set.     

Flexible uncertainty in mixture models for ordinal responses

In classical mixture models for ordinal data with an uncertainty component, the Uniform distribution is used to model indecision. In the approach proposed here, the discrete Uniform distribution is replaced by a more flexible distribution, which is centered in the middle of the response categories. The resulting model allows to distinguish between a tendency to middle categories and a tendency to extreme categories. By linking these preferences to explanatory variables, one can investigate which persons show a tendency to these response styles. It is demonstrated that severe bias might occur if inadvertently the Uniform distribution is used to model uncertainty. An application to attitudes on the performance of health services illustrates the advantages of the more flexible model.     

Rationalizing Policy Functions by Dynamic Optimization

We derive necessary and sufficient conditions for a pair of functions to be the optimal policy function and the optimal value function of a dynamic maximization problem with convex constraints and concave objective functional. It is shown that every Lipschitz continuous function can be the solution of such a problem. If the maintained assumptions include free disposal and monotonicity, then we obtain a complete characterization of all optimal policy and optimal value functions. This is the case, e.g., in the standard aggregative optimal growth model.