Permutation Methods for Comparing the Accuracy of Nested Prediction Models in Survival Analysis

When making patient-specific prediction, it is important to compare prediction models to evaluate the gain in prediction accuracy for including additional covariates. We propose two statistical testing methods, the complete data permutation (CDP) and the permutation cross-validation (PCV) for comparing prediction models. We simulate clinical trial settings extensively and show that both methods are robust and achieve almost correct test sizes; the methods have comparable power in moderate to large sample situations, while the CDP is more efficient in computation. The methods are also applied to ovarian cancer clinical trial data.     

Model-based adaptive spatial sampling for occurrence map construction

In many environmental management problems, the construction of occurrence maps of species of interest is a prerequisite to their effective management. However, the construction of occurrence maps is a challenging problem because observations are often costly to obtain (thus incomplete) and noisy (thus imperfect). It is therefore critical to develop tools for designing efficient spatial sampling strategies and for addressing data uncertainty. Adaptive sampling strategies are known to be more efficient than non-adaptive strategies. Here, we develop a model-based adaptive spatial sampling method for the construction of occurrence maps. We apply the method to estimate the occurrence of one of the world’s worst invasive species, the red imported fire ant, in and around the city of Brisbane, Australia. Our contribution is threefold: (i) a model of uncertainty about invasion maps using the classical image analysis probabilistic framework of Hidden Markov Random Fields (HMRF), (ii) an original exact method for optimal spatial sampling with HMRF and approximate solution algorithms for this problem, both in the static and adaptive sampling cases, (iii) an empirical evaluation of these methods on simulated problems inspired by the fire ants case study. Our analysis demonstrates that the adaptive strategy can lead to substantial improvement in occurrence mapping.     

Comparison of estimators in nested case–control studies with multiple outcomes

Reuse of controls in a nested case-control (NCC) study has not been considered feasible since the controls are matched to their respective cases. However, in the last decade or so, methods have been developed that break the matching and allow for analyses where the controls are no longer tied to their cases. These methods can be divided into two groups; weighted partial likelihood (WPL) methods and full maximum likelihood methods. The weights in the WPL can be estimated in different ways and four estimation procedures are discussed. In addition, we address modifications needed to accommodate left truncation. A full likelihood approach is also presented and we suggest an aggregation technique to decrease the computation time. Furthermore, we generalize calibration for case-cohort designs to NCC studies. We consider a competing risks situation and compare WPL, full likelihood and calibration through simulations and analyses on a real data example.     

Effect of an Accessibility Measure in a Model for Choice of Residential Location,Workplace, and Type of Employment

A three-level nested logit model for the choice of residential location, workplace, and type of employment is used to assess the effect of an individual-specific measure of accessibility to employments that takes into account the attractiveness of different occupations when the choice of workplace is anticipated in the decision of residential location. The model allows for variation in the preferences for types of employment across individuals and accounts for individual heterogeneity of preferences at each choice level in education, age, gender, and children. Using data from the Île-de-France region, the model shows that the individual-specific accessibility measure is an important determinant of the choice of residential location and its effect differs along the life cycle. The attractiveness of the types of employment is a better predictor of the workplace location than the usual total number of employments.     

Social upgrading in globalized production: The case of the textile and clothing industry

Vertical specialization generated by the international fragmentation of production within global networks is driven not only by comparative advantage, but also by the locational decisions of lead firms which determine the role and bargaining power of local producers in their value chain. This study examines the consequences of such specialization in textiles and clothing for 26 labour‐abundant countries from 1990 to 2007. Fixed effects regressions based on panel data reveal that the industry does not always reap the benefits of the resulting international trade integration. Rather, the authors observe a negative relationship between vertical specialization and relative real wages in the textile and clothing industry.     

COMMUNITY SERVICES AND OUT-MIGRATION

This paper investigates the relationship between changing community context and out-migration in one of today's poor countries, seeking to document the various mechanisms by which infrastructure affects the migratory behavior. We focus on the expansion of social and physical facilities and services near to rural people's homes, including transportation, new markets, employment, schools, health clinics, and mass media outlets such as movie halls. We draw upon detailed data from Nepal to estimate the hypothesized effects. The direct effects of expanding economic and human capital infrastructure are clearly negative, reducing out-migration. However, increased economic infrastructure is associated with a greater accumulation of human and social capital among respondents and their parents. Through these intervening mechanisms, economic and social infrastructure increased the odds of migrating out. These results reveal the often countervailing nature of short- and long-term effects of economic and social change and the complex pathways influencing migration outcomes.     

The Kaleidoscope of the Sociology of Aging Seen through the French-Speaking Field

How does French-language sociology deal with aging? This is the question posed by this contribution, which explores articles published within the past fifteen years in various French-language scientific journals. Through a systematic review of the scientific literature, we analyze the evolution of topics in the sociology of aging through five well-known French-language journals of sociology. Three of them are generalist journals (Revue française de sociologie, Revue suisse de sociologie, Recherches sociologiques et anthropologiques), from France, Switzerland, and Belgium respectively. The remaining two journals, both from France, are specifically devoted to the study of aging (Gérontologie et société, Retraite et société). The results invite further exploration of the influence of the national contexts in which French-speaking European sociologies are structured.     

I’m not like everybody else: urbanization factors shaping spatial distribution of native and invasive ants are species-specific

Urbanization is a major global change inducing complex and multiple modifications of landscapes and ecosystems. The spatial distributions of organisms experiencing these modifications will likely shift specifically, depending on each species’ response to each environmental modification induced by urbanization. We sampled two ant genera (Lasius and Tetramorium) at 1248 locations along an urbanization gradient in Lyon, France and used high resolution spatial layers for 18 spatial (e.g., open habitat fragmentation, bioclimatic data and surface temperatures) and temporal (e.g., comparison of Normalized Difference Vegetation Index between 1986 and 2015) environmental variables associated with urbanization. Coupling two different analytical methods (Outlying Mean Index and Boosted Regression Trees), we showed that each species’ distribution was influenced by its own combination of environmental factors. Two morphologically cryptic Tetramorium species (T. sp.E and T. sp.U2) were both highly abundant but with opposite responses to urbanization: while T. sp.E was favored by urbanized habitat, T. sp.U2 avoided urbanized areas. Among Lasius species, we detected 63 occurrences of the invasive ant Lasius neglectus, the distribution of which was favored only by embankments along roads. We found that, even at this reduced spatial scale, climatic effects influenced most species and interacted with urbanization factors.     

Measuring educational inequalities: a method and an application to Albania

In this paper, we investigate whether educational inequalities stem from differences between families or within families. In a poor economy, schooling is costly for parents, and education is likely to be unequally distributed among siblings. Drawing on discrete ordered choice models, we present a simple method to estimate the between and within components of both the explained and unexplained variance in education. For our empirical analysis, we use the Living Standard Measurement Study survey conducted in 2002 in Albania. We explain about 40% of the total variance and find that inequalities in education are mainly due to differences between families. Differences within families are smaller and are far less easily explained.