A data-driven selection of the number of clusters in the Dirichlet allocation model via Bayesian mixture modelling

In this paper, we consider a Bayesian mixture model that allows us to integrate out the weights of the mixture in order to obtain a procedure in which the number of clusters is an unknown quantity. To determine clusters and estimate parameters of interest, we develop an MCMC algorithm denominated by sequential data-driven allocation sampler. In this algorithm, a single observation has a non-null probability to create a new cluster and a set of observations may create a new cluster through the split-merge movements. The split-merge movements are developed using a sequential allocation procedure based in allocation probabilities that are calculated according to the Kullback–Leibler divergence between the posterior distribution using the observations previously allocated and the posterior distribution including a ‘new’ observation. We verified the performance of the proposed algorithm on the simulated data and then we illustrate its use on three publicly available real data sets.     

Bhattacharyya and Kshirsagar Bounds in Generalized Gamma Distribution

In this article, we consider the exact computation of the famous halfspace depth (HD) and regression depth (RD) from the view of cutting a convex cone with hyperplanes. Two new algorithms are proposed for computing these two notions of depth. The first one is relatively straightforward but quite inefficient, whereas the second one is much faster. It is noteworthy that both of them can be implemented to spaces with dimension beyond three. Some numerical examples are also provided in what follows to illustrate the performances.     

Generalizing the Bayesian Vector Autoregression Approach for Regional Interindustry Employment Forecasting

The Bayesian vector autoregression (BVAR) employment-forecasting approach is generalized using data for the state of Georgia. This study advances previous regional BVAR approaches by (a) incorporating regional input-output coefficients instead of national coefficients, (b) using the coefficients both to specify the prior means in one model and to weight the variances of a Minnesota-type prior in a second model, and (c) including final-demand effects and links to national and world economies. Out-of-sample forecasts produced by the generalized BVAR models are compared to forecasts produced from an autoregressive model, an unconstrained VAR model, and a Minnesota BVAR model.     

Migration to London and the development of the north–south divide, 1851–1911

This article uses census data for England and Wales covering the period 1851–1911 to provide new insights into patterns of migration to London. It examines several related themes including the role migration played in London’s growth during this period, age and gender differentials and distance travelled. Calculating net migration rates, the article demonstrates that after age 30, of those born outside of London, more left the Capital than came, yet over time an increasing proportion of the migrant population was retained. The proportion of family migrants fluctuated over the period, yet compared to others tended to travel shorter distances, a feature which increased over time with suburbanization. Turning to the geographical origins of migrants, London drew migrants from across the entirety of England and Wales. However, the data suggest that the migrant sex ratio became more homogeneous over time, with distinct pockets of male dominated migration that were visible in 1851 disappearing by 1911. Lastly, the article investigates migration from the perspective of place of departure rather than destination, as is traditionally the case. This reveals a distinct regional geography, suggesting that the present-day north–south divide was already evident in 1851, and became increasingly distinct over time.     

Reliability and validity of the Japanese version of the Survey Work–Home Interaction – NijmeGen,the SWING (SWING-J)

ABSTRACT

This study sought to validate the Japanese version of the Survey Work–Home Interaction – NijmeGen, the SWING, which assesses multi-dimensional work–family interaction by differentiating between the direction and quality of influence. We translated the SWING into the Japanese language, the SWING-J. A back-translation procedure confirmed that the translation was appropriate. A total of 2701 dual-earner parents with preschool children (1193 men and 1508 women) were surveyed. The complete questionnaire included the SWING-J, job and family domain variables, and well-being indicators. The reliability and factorial and convergent validity of the used measures were examined. As the results, four dimensions (i.e. work-to-family negative spillover, family-to-work negative spillover, work-to-family positive spillover and family-to-work positive spillover) were determined by an exploratory factor analysis. A series of confirmatory factor analyses suggested that the hypothesized four-factor model provided a reasonably good fit to the data. Convergent validity was generally supported by the expected correlations of work–family spillovers with the possible predictors and consequences. Cronbach's alpha coefficients of the four subscales of the SWING-J were satisfactory (0.75–0.86). The present study confirmed that the Japanese version of the SWING is an adequate tool to measure positive and negative spillover between working life and family life among Japanese workers.     

Exploring the role of gain versus loss framing and point of reference in messages to reduce human–bear conflicts

As a fundamental social process, communication can help natural resource managers prevent or manage many human–wildlife conflicts, but research on how to present effectively information that contributes to the reduction of human–wildlife conflicts is scant. We examined the effectiveness of point-of-reference and gain-versus-loss framing in heightening intentions to prevent human–black bear conflicts. We randomly assigned 811 participants to one of six message conditions as part of a 3 (point of reference: family versus community versus bear) × 2 (message framing: gain versus loss) between-subjects factorial design or a control condition. The findings show that a match between loss framing and low-construal referencing point (e.g., family-referencing), and a congruency between gain framing and high-construal referencing point (e.g., bear-referencing) lead to higher risk perception and behavioral intentions. This study provides guidance for continuing research and design of risk communication in the context of human–bear conflicts.     

Neither global nor local: Heterogeneous connectivity in spatial network structures of world migration

For a long time, geographic regions were considered the dominant spatial arbiter of international migration of people. However, since the late 1970s, many scholars have argued that movements reach beyond contiguous regions to connect distant, dispersed, and previously disconnected countries across the globe. The precise structure of world migration, however, remains an open question. We apply network analysis that incorporates spatial information to international migration-stock data to examine what multilateral structures of world migration have emerged from the interplay of regional concentration (local cohesion) and global interconnectedness (global cohesion) for the period 1960–2000. In the world migration network (WMN), nodes represent countries located in geographic space, and edges represent migrants from an origin country who live in a destination country during each decade. We characterize the large-scale structure and evolution of the WMN by algorithmically detecting international migration communities (i.e., sets of countries that are densely connected via migration) using a generalized modularity function for spatial, temporal, and directed networks. Our findings for the whole network suggest that movements in the WMN deviate significantly from the regional boundaries of the world and that international migration communities have become globally interconnected over time. However, we observe a strong variability in the distribution of strengths, neighborhood overlaps, and lengths of migration edges in the WMN. This manifests as three types of communities: global, local, and glocal. We find that long-distance movements in global communities bridge multiple non-contiguous countries, whereas local (and, to a lesser extent, glocal) communities remain trapped in contiguous geographic regions (or neighboring regions) for almost the whole period, contributing to a spatially fragmented WMN. Our findings demonstrate that world migration is neither regionally concentrated nor globally interconnected, but instead exhibits a heterogeneous connectivity pattern that channels unequal migration opportunities across the world.     

EDITORIAL

Although a contested concept, there is a growing awareness of the phenomenon of deliberate self‐harm (DSH) in the UK, and particularly of self‐injury by adolescents. Trends in the prevalence of DSH have been monitored, and schools have been used as a means for gathering self‐reported data from pupils, but to‐date there is no known research on the way self‐harm is perceived and responded to by schools and the agencies to which they refer. This paper reports a Nuffield Foundation‐funded project which examined the experiences and perceptions of DSH of 34 teachers and others in support and clinical roles. The report includes data on the range and prevalence of self‐harming behaviours encountered; levels of awareness amongst school staff; teachers' reactions to DSH; links between schools and other agencies (including CAMHS); and levels of training, support and supervision for teachers and others responding to DSH in school settings. It concludes with a discussion of the implications of the findings, including the potential for CAMHS, social workers and other agencies to provide crucial training and support to ‘front‐line workers’ in schools.