On predictive distributions and Bayesian networks

In this paper we are interested in discrete prediction problems for a decision-theoretic setting, where the task is to compute the predictive distribution for a finite set of possible alternatives. This question is first addressed in a general Bayesian framework, where we consider a set of probability distributions defined by some parametric model class. Given a prior distribution on the model parameters and a set of sample data, one possible approach for determining a predictive distribution is to fix the parameters to the instantiation with the maximum a posteriori probability. A more accurate predictive distribution can be obtained by computing the evidence (marginal likelihood), i.e., the integral over all the individual parameter instantiations. As an alternative to these two approaches, we demonstrate how to use Rissanen's new definition of stochastic complexity for determining predictive distributions, and show how the evidence predictive distribution with Jeffrey's prior approaches the new stochastic complexity predictive distribution in the limit with increasing amount of sample data. To compare the alternative approaches in practice, each of the predictive distributions discussed is instantiated in the Bayesian network model family case. In particular, to determine Jeffrey's prior for this model family, we show how to compute the (expected) Fisher information matrix for a fixed but arbitrary Bayesian network structure. In the empirical part of the paper the predictive distributions are compared by using the simple tree-structured Naive Bayes model, which is used in the experiments for computational reasons. The experimentation with several public domain classification datasets suggest that the evidence approach produces the most accurate predictions in the log-score sense. The evidence-based methods are also quite robust in the sense that they predict surprisingly well even when only a small fraction of the full training set is used.     

Putting Process (Back) In: Research on the Internationalization Process of the Firm

This paper provides a critique of existing research on the internationalization process of the firm and proposes an agenda for future inquiry. In recent years, process approaches have received increasing attention in management research, leading to a more refined understanding of the distinction between process and variance paradigms. We apply a process lens to a well‐established sub‐field of international business, namely the internationalization process of the firm. We review how this research tradition has evolved over four decades. The review commences with a reassessment of the seminal ‘stage models’ that date back to the 1970s. It then proceeds to classify subsequent research on the basis of whether it includes process data and/or process theorizing. It is found that the majority of studies in this review do not combine process data with process theorizing. We show how, even in studies that contain some process elements, a process approach is not always sustained throughout the paper. On the basis of this review, six research themes are proposed, which would form the basis for a process agenda for future research.     

Multiple hypothesis tests based On conditional differences in means

Many hypothesis tests are univariate tests and cannot cope with multiple hypothesis without an auxiliary procedure as e. g. the Bonferroni-Holm-procedure. At the same time, there is an urgent need for testing multiple hypothesis due to the very simple existing methods as the Bonferroni-correction or the Bonferroni-Holm-procedure, which suffers from a very small local significance level to detect statistical inferences or the drawback that logical and statistical dependencies among the test statistics are not used, whereby its detection is NP-hard. In honour of this occasion, we present a multiple hypothesis test for i.i.d. random variables based on conditional differences in means, which is capable to cope with multiple hypothesis and does not suffer on such drawbacks as the Bonferroni-correction or the Bonferroni-Holm-procedure. Thereby, the computation time can be neglected.     

Probabilistic solutions to ordinary differential equations as nonlinear Bayesian filtering: a new perspective

Statistics and Computing - We formulate probabilistic numerical approximations to solutions of ordinary differential equations (ODEs) as problems in Gaussian process (GP) regression with nonlinear...     

Participation in Everyday Life Before and After a Housing Adaptation

Housing adaptation aims to enable clients to live independently in their own homes. Studies focusing on participation in everyday life following a housing adaptation are lacking and needed. This study aimed to explore housing adaptation clients' experiences of participation in everyday life before and after a housing adaptation, through the lens of a housing adaptation, using a qualitative follow-up design, with 11 participants. It was found that when the housing adaptation met the participants' needs, performance of activities improved and the housing adaptation opened doors to engagement and participation in everyday life. Thus, focus on performance and engagement in everyday life at the onset of the housing adaptation process, combined with regular follow-ups, may enhance participation.     

Education,cognitive ability,and cause-specific mortality: A structural approach

Education is negatively associated with most major causes of death. Prior work ignores the premise that cause-specific hazards are interdependent and that both education and mortality depend on cognitive ability. We analyse Swedish men aged 18–63, focusing on months lost due to specific causes—which solves the interdependence problem—and use a structural model that accounts for confounding due to cognitive ability. In a standard Cox model controlling for Intelligence Quotient, improving education is associated with large decreases in mortality for major causes of death. In the structural model, improving education is associated with a small decrease in months lost for most causes and education levels. Among the least educated, however, improving education strongly reduces the months lost, mainly those lost from external causes, such as accidents and suicide. Results suggest that conventional analysis of education and mortality may be biased, even if accounting for observed cognition.     

Forced Migration and Mortality in the Very Long Term: Did Perestroika Affect Death Rates Also in Finland?

In this article, we analyze mortality rates of Finns born in areas that were ceded to the Soviet Union after World War II and from which the entire population was evacuated. These internally displaced persons are observed during the period 1971-2004 and compared with people born in the same region but on the adjacent side of the new border. We find that in the 1970s and 1980s, the forced migrants had mortality rates that were on par with those of people in the comparison group. In the late 1980s, the mortality risk of internally displaced men increased by 20% in relation to the expected time trend. This deviation, which manifests particularly in cardiovascular mortality, coincides with perestroika and the demise of the Soviet Union, which were events that resulted in an intense debate in civil society about restitution of the ceded areas. Because state actors were reluctant to engage, the debate declined after some few years, and after the mid-1990s, the death risk again approached the long-term trend. Our findings indicate that when internally displaced persons must adjust to situations for which appropriate coping behaviors are unknown, psychosocial stress might arise several decades after their evacuation.     

Divorce and parity progression following the death of a child: A register-based study from Finland

Most studies that have examined whether a child’s death influences parental relationship stability have used small-scale data sets and their results are inconclusive. A likely reason is that child loss affects not only the risk of parental separation, but also the risk of having another child. Hence parity progression and separation must be treated as two competing events in relation to child loss. The analysis in this paper used Finnish register data from 1971 to 2003, covering over 100,000 married couples whose durations of both first marriage and parenthood could be observed. We ran parity-specific Cox regressions in which process time started from the birth of each additional child. All marriages included women of childbearing age, none of whom had experienced any child death on entering the analysis. We find that child loss only modestly influences the divorce risk, whereas its effect on the risk of parity progression is considerable.     

Die Finanzialisierung von Unternehmen

Since the 1990s the financialization of the enterprise is an increasingly important aspect of capitalist development also in Germany. However, more recently a growing number of scholars point at a problematic conceptual stretching of financialization. The term has been defined too differently and ambiguously and has not been distinguished sufficiently from seemingly similar concepts. Financialization has been prematurely identified with a set of fixed effects often condensed to stagist concepts of capitalist development. When analyzing effects financialization has been regarded in isolation without considering interaction effects with other structurations of fields and hence the multi-referentiality of the enterprise. In order to overcome these shortcomings the article proposes a concept for the analysis of firm level financialization which draws on a more general sociological concept of field structuration and dynamics in which the “multi-referential enterprise” operates. Based on this concept the article analyses episodes of financialization in Germany showing its dynamics and limits. Financialization of enterprises proves to be structurally limited, institutionally contested, and unfolds its effects in deferent actor constellations at firm level due to its multiple and contradictory embeddedness.     

CHAOTIC DYNAMICS IN CONTESTS

Under a myopic best‐reply dynamic, efforts in repeated contests may exhibit chaotic behavior. This may help explain, for example, why experimental data often show nonconvergence to one‐shot equilibrium efforts. (JEL C61, C72, D72, D74, D83)