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.     

Radical Distrust: Are Economic Policy Attitudes Tempered by Social Trust?

Social Indicators Research - Debates about the appropriate role of markets and governments are often shaped by sharply contrasting opinions. Based on individual data from the World Values Survey...     

A vector of quarters representation for bivariate time series

In this paper it is shown that several models for a bivariate nonstationary quarterly time series are nested in a vector autoregression with cointegration restrictions for the eight annual series of quarterly observations. Or, the Granger Representation Theorem is extended to incorporate, e.g., seasonal and periodic cointegration.     

Book reviews

Statistical Papers -     

Detecting seasonal unit roots in a structural time series model

In this paper, we propose to detect seasonal unit roots within the context of a structural time series model. Such a model is often found to be useful in practice. Using Monte Carlo simulations, we show that our method works well. We illustrate our approach for several quarterly macroeconomic time series variables.     

Public pensions in transition

The main purpose of this paper is to analyze problems of financing an old-age insurance when birth rates are low and population declines or fertility fluctuates with time. A government then searches for optimal policies to cope with such problems. A first criterion could be seen in the Pareto principle. But we all know that there is no way out of PAYG unless at least one generation has to pay for the transition. Therefore an optimal policy is concerned with intergenerational redistribution and optimal growth.In the absence of public pensions the economy will in the long run converge to a steady state which is not optimal in the sense of a golden rule. This dynamic in-efficiency results from the decentralized decision making by the consumers and the firms. If the PAYG system influences the savings ratio of the economy, public pensions can be seen as an instrument to implement a modified golden rule.Paper presented at the ISPE-conference on The Fiscal Implications of an Ageing Population, Vaalsbroek, The Netherlands, May 30–June 1, 1990. I am obliged to Dieter Bös, Friedrich Breyer, Christian Keuschnigg, Wolfgang Kitterer, and an anonymous referee for helpful comments on an earlier draft of this paper. Financial support through Deutsche Forschungsgemeinschaft, SFB 303 is gratefully acknowledged.     

Hierarchy or co-operation: labour-management relations in church institutions

Aspects of industrial relations and internal co-ordination have been a major topic of empirical research in the German business enterprise sector. In contrast, there are practically no empirical investigations into labour-management relations in third sector institutions. The largest share of this sector in Germany is represented by the two Christian churches with presently about 900,000 employees. According to the notion of church officials, labour-management relations in these institutions are characterised by a specific normative concept which emphasises a trustful co-operation of all employees in the mission of the church. The major concern of our research project focused on the question whether the specific conditions of employment within church institutions favour more hierarchical or more co-operative forms of co-ordination and how employees and employee representatives perceive the ideal of the community of service (Dienstgemeinschaftsgedanke). The results of our research lead to the conclusion that church labour-management relations are characterised by a remarkable contrast between the participative or co-operative notion of the ideal of the community of service and present forms of hierarchical co-ordination inside church institutions.This article gives a selected overview of Beyer and Nutzinger (1991). The research was supported by the Hans-Böckler-Stiftung (Düsseldorf) and the University of Kassel.     

The Determinants of Trust and Credibility in Environmental Risk Communication: An Empirical Study

This study examines a key component of environmental risk communication; trust and credibility. The study was conducted in two parts. In the first part, six hypotheses regarding the perceptions and determinants of trust and credibility were tested against survey data. The hypotheses were supported by the data. The most important hypothesis was that perceptions of trust and credibility are dependent on three factors: perceptions of knowledge and expertise; perceptions of openness and honesty; and perceptions of concern and care. In the second part, models were constructed with perceptions of trust and credibility as the dependent variable. The goal was to examine the data for findings with direct policy implications. One such finding was that defying a negative stereotype is key to improving perceptions of trust and credibility.