Improved Transferability of Data‐Driven Damage Models Through Sample Selection Bias Correction

Damage models for natural hazards are used for decision making on reducing and transferring risk. The damage estimates from these models depend on many variables and their complex sometimes nonlinear relationships with the damage. In recent years, data‐driven modeling techniques have been used to capture those relationships. The available data to build such models are often limited. Therefore, in practice it is usually necessary to transfer models to a different context. In this article, we show that this implies the samples used to build the model are often not fully representative for the situation where they need to be applied on, which leads to a “sample selection bias.” In this article, we enhance data‐driven damage models by applying methods, not previously applied to damage modeling, to correct for this bias before the machine learning (ML) models are trained. We demonstrate this with case studies on flooding in Europe, and typhoon wind damage in the Philippines. Two sample selection bias correction methods from the ML literature are applied and one of these methods is also adjusted to our problem. These three methods are combined with stochastic generation of synthetic damage data. We demonstrate that for both case studies, the sample selection bias correction techniques reduce model errors, especially for the mean bias error this reduction can be larger than 30%. The novel combination with stochastic data generation seems to enhance these techniques. This shows that sample selection bias correction methods are beneficial for damage model transfer.     

Imposing and Testing for Shape Restrictions in Flexible Parametric Models

In many economic models, theory restricts the shape of functions, such as monotonicity or curvature conditions. This article reviews and presents a framework for constrained estimation and inference to test for shape conditions in parametric models. We show that “regional” shape-restricting estimators have important advantages in terms of model fit and flexibility (as opposed to standard “local” or “global” shape-restricting estimators). In our empirical illustration, this is the first article to impose and test for all shape restrictions required by economic theory simultaneously in the “Berndt and Wood” data. We find that this dataset is consistent with “duality theory,” whereas previous studies have found violations of economic theory. We discuss policy consequences for key parameters, such as whether energy and capital are complements or substitutes.     

Optimization problems with color-induced budget constraints

In 1984, Gabow and Tarjan provided a very elegant and fast algorithm for the following problem: given a matroid defined on a red and blue colored ground set, determine a basis of minimum cost among those with k red elements, or decide that no such basis exists. In this paper, we investigate extensions of this problem from ordinary matroids to the more general notion of poset matroids which take precedence constraints on the ground set into account. We show that the problem on general poset matroids becomes -hard, already if the underlying partially ordered set (poset) consists of binary trees of height two. On the positive side, we present two algorithms: a pseudopolynomial one for integer polymatroids, i.e., the case where the poset consists of disjoint chains, and a polynomial algorithm for the problem to determine a minimum cost ideal of size l with k red elements, i.e., the uniform rank-l poset matroid, on series-parallel posets.     

Risk of Extreme Events in Multiobjective Decision Trees Part 1. Severe Events

Earlier work with decision trees identified nonseparability as an obstacle to minimizing the conditional expected value, a measure of the risk of extreme events, by the well-known method of averaging out and folding back. This first of two companion papers addresses the conditional expected value that is defined as the expected outcome assuming the exceedance of a threshold β, where β is preselected by the decision maker. An approach is proposed to overcome the need to evaluate all policies in order to identify the optimal policy. The approach is based on the insight that the conditional expected value is separable into two constituent elements of risk and can thus be optimized along with other objectives, including the unconditional expected value of the outcome, by using a multiobjective decision tree. An example of sequential decision making for improving highway capacity is given.     

The Distribution of the German Public–Private Wage Gap

This paper provides a distributional analysis of the public–private sector wage gap in Germany from 1984 to 1996. The public sector wage distribution is generally less dispersed than the private sector wage distribution. The raw wage differential is positive for males who are at the lower tail of the male wage distribution and negative at the upper tail. In contrast, females enjoy positive wage gaps along most part of the wage distribution. A decomposition analysis reveals that the male wage premium, i.e. the part of the wage gap not accounted for by differences in observable characteristics, is uniformly negative, whereas the female wage premium is positive.     

How Managers Talk about their Consumption of Popular Management Concepts: Identity,Rules and Situations

This paper analyses how top managers account for their consumption of popular management concepts. By ‘consumption’ we refer to managers acting as active users of popular management concepts within their organizations. After reviewing the relevant literature, we argue that the logic of appropriateness is a better theoretical perspective to view, understand and analyse managers' accounts of concept consumption than is the logic of consequence. We apply this perspective to extensive interviews we conducted with top managers in Germany. Based on the managers' own accounts of how they understand and apply popular management concepts, we identified four discourse categories: (1) learning from others' experiences, (2) controlling organizational change, (3) gaining external legitimacy and (4) collective sensemaking. We argue that these discourse categories all draw on the social norm of rationality central to managerial identity, while differing in socially defined rules about how rationality is realized in typical management situations. Our findings strongly encourage researchers, when investigating popular management concepts in the future, to take into account the situational nature of rationality that circumstantiates the consumption of concepts.     

Upgrading local operations for global arbitrage

This study shows that upgrading sales operations in host countries towards production allows MNCs to perform global arbitrage in their networks of foreign subsidiaries and thereby contributes to a strategy of reaping the benefits of multinationality. We predict and find that operation upgrades follow opportunities of improving resource flows in the subsidiary network to exploit the advantages of host countries in tax rates, investment incentives, and technological knowledge across borders. Performance effects on the level of the MNC network give evidence of these benefits from global arbitrage.     

Influencing alcohol control policies and practices at community festivals

The goal of this study was to assess the feasibility and effectiveness of two interventions aimed at reducing alcohol-related risks at community festivals--a training program for festival planners and a community organizing campaign. We randomly selected four festivals for each intervention and had 24 comparison festivals. Our assessment included process evaluation to track and evaluate types of alcohol policies resulting from each of the interventions, pre and post telephone surveys of key festival planners and law enforcement agencies, and pre and post pseudo-underage and pseudo-intoxicated purchase attempts. Analyses showed that both interventions were feasible and were successful in influencing adoption of written policies and improving alcohol-related practices. However, neither intervention appeared to decrease propensity for illegal alcohol sales at these events, likely due, in part, to the short time frame of the interventions. Future research should assess effects of the interventions on alcohol-related problems and effects of enforcement interventions.     

The effect of compulsory schooling on health—evidence from biomarkers

Using data from the Health Survey for England and the English Longitudinal Study on Ageing, we estimate causal effects of schooling on health. Our study complements earlier studies exploiting two nationwide increases in British compulsory school leaving age in 1947 and 1973, respectively, by using biological stress markers as measures of health outcomes in addition to self-reported measures. We find a strong positive correlation between education and health, both self-rated and measured by blood fibrinogen and C-reactive protein levels. However, causal effects estimates based on compulsory schooling changes are ambiguous and remain statistically insignificant.