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.     

Learning from a lot: Empirical Bayes for high‐dimensional model‐based prediction

Empirical Bayes is a versatile approach to “learn from a lot” in two ways: first, from a large number of variables and, second, from a potentially large amount of prior information, for example, stored in public repositories. We review applications of a variety of empirical Bayes methods to several well‐known model‐based prediction methods, including penalized regression, linear discriminant analysis, and Bayesian models with sparse or dense priors. We discuss “formal” empirical Bayes methods that maximize the marginal likelihood but also more informal approaches based on other data summaries. We contrast empirical Bayes to cross‐validation and full Bayes and discuss hybrid approaches. To study the relation between the quality of an empirical Bayes estimator and p, the number of variables, we consider a simple empirical Bayes estimator in a linear model setting. We argue that empirical Bayes is particularly useful when the prior contains multiple parameters, which model a priori information on variables termed “co‐data”. In particular, we present two novel examples that allow for co‐data: first, a Bayesian spike‐and‐slab setting that facilitates inclusion of multiple co‐data sources and types and, second, a hybrid empirical Bayes–full Bayes ridge regression approach for estimation of the posterior predictive interval.     

Standing on Their Own Two Feet: How the New Public Charge Rules Could Impact Non-European LPR Applicants

Population Research and Policy Review - In February 2020, the U.S. government began to implement a new Public Charge rule that greatly expands the definition of “public charge” when...     

The Welfare State as Safety Net in Migration Preferences: Empirical Evidence from an Experiment Among Dutch Master Students

Population Research and Policy Review - The welfare state can be perceived as a safety net which helps individuals adjust to situations of risk or transition. Starting from this idea of the welfare...     

Inequality Assessment by Probabilistic Development Indices

Social Indicators Research - This paper analyses the Human Development Index (HDI) time series from 2010 to 2017. An alternative index is studied, which combines the same components of the HDI by...     

Leader-employee congruence of expected contributions in the employee-organization relationship

Employees' expected contributions can be incongruent with those of their leader. We examine the congruence effect of leaders' and employees' expected contributions on job satisfaction. Results of cross-level polynomial regressions on 947 employees and 224 leaders support the congruence effect. When expected contributions are congruent, employees are more satisfied with their job. Our findings suggest that employees enjoy high challenges, as long as these challenges are in harmony with the expected contributions of their leaders. Employees are less satisfied with their jobs both when their expected contributions were higher than their leaders' and when their expected contributions were lower than those of their leaders. Beyond the relevance of having high expected contributions, the findings highlight the crucial role played by the congruence of expected contributions of leaders and employees.     

Myriad potential for mentoring: Understanding the process of transformational change through a gender equality intervention

This article contributes to understanding transformational change towards gender equality by examining the transformational change potential of a mentoring programme for women, a type of gender equality intervention both criticized and praised for its ability to bring about change. Drawing upon an empirical case study of a mentoring programme for women academics in a Dutch university, we explore three dimensions of transformational change: organizational members (i) discussing and reflecting upon gendered organizational norms and work practices; (ii) creating new narratives; and (iii) experimenting with new work practices. Our findings indicate five specific conditions that enable transformational change: cross‐mentoring, questioning what is taken for granted, repeating participation and individual stories, facilitating peer support networks and addressing and equipping all participants as change agents. We suggest that these conditions should be taken into account when (re)designing effective organizational gender equality interventions.     

Journey towards independent living: a grounded theory investigation of leaving the care of Girls & Boys Town,South Africa

The journey out of care and towards independent living is a challenge for many care-leavers. There has been little research into the social processes involved in this care-leaving journey. This paper presents the results of a grounded theory investigation into the care-leaving journeys of nine young men who had, several years previously, been in the care of Girls & Boys Town in South Africa. Working from a resilience perspective, with an ecological emphasis, four central social processes emerged that together explain the care-leaving experiences of the participants. These processes are striving for authentic belonging; networking people for goal attainment; contextualised responsiveness and building hopeful and tenacious self-confidence. These four processes are located within contextual boundaries and at the social environmental interface. The paper presents these processes in detail, drawing on selected narratives of the participants and integrated with additional theory. It is hoped that this paper may contribute to theory building concerning care-leaving processes and enhance youth care practices for youth in care and leaving care.