Compositional biplots including external non-compositional variables

Biplots represent a widely used statistical tool for visualizing the resulting loadings and scores of a dimension reduction technique applied to multivariate data. If the underlying data carry only relative information (i.e. compositional data expressed in proportions, mg/kg, etc.) they have to be pre-processed with a logratio transformation before the dimension reduction is carried out. In the context of principal component analysis, the resulting biplot is called compositional biplot. We introduce an alternative, the ilr biplot, which is based on a special choice of orthonormal coordinates resulting from an isometric logratio (ilr) transformation. This allows to incorporate also external non-compositional variables, and to study the relations to the compositional variables. The methodology is demonstrated on real data sets.     

Structure of Fibonacci cubes: a survey

The Fibonacci cube Γ _n is the subgraph of the n-cube induced by the binary strings that contain no two consecutive 1s. These graphs are applicable as interconnection networks and in theoretical chemistry, and lead to the Fibonacci dimension of a graph. In this paper a survey on Fibonacci cubes is given with an emphasis on their structure, including representations, recursive construction, hamiltonicity, degree sequence and other enumeration results. Their median nature that leads to a fast recognition algorithm is discussed. The Fibonacci dimension of a graph, studies of graph invariants on Fibonacci cubes, and related classes of graphs are also presented. Along the way some new short proofs are given.     

Current Developments in Social Care Services for Older Adults in the Czech Republic: Trends Towards Deinstitutionalization and Marketization

ABSTRACT

Deinstitutionalization and marketization of eldercare has been delayed by 20 years in the Czech Republic compared to Western Europe, and it takes place in a completely different context, characterized by the legacy of communism, a growing older population, and less generous public subsidies. This study is the first in the Czech Republic to examine how deinstitutionalization and marketization effect implementation of these principles on the availability and quality of eldercare services at regional and municipal levels. A mixed-method approach was used, combining several data sources (policy documents, administrative data, statistics, expert panel, and secondary use of qualitative data). The findings suggest that the support for and availability of home-based care has declined, despite the ever-increasing number of older adults and policy preference for deinstitutionalization. Furthermore, home-based services have failed to adjust to growing care needs of older adults (e.g., inflexible schedules, limited provision of time-demanding care, inadequate staff composition). This situation occasioned an unintended outcome: the emergence of nonregistered, semilegal, for-profit nursing homes offering low-quality care and poor working conditions, and subject to no quality control. The health and even lives of older adults are at risk if they choose such services. Research is needed to study older adult decision making and offer them tools to identify and avoid questionable services.     

Core/periphery scientific collaboration networks among very similar researchers

Empirical studies such as Goyal et al. (J Polit Econ 114(2):403–412, 2006) or Newman (Proc Natl Acad Sci USA 101(Suppl. 1):5200–5205, 2004) show that scientific collaboration networks present a highly unequal and hierarchical distribution of links. This implies that some researchers can be much more active and productive than others and, consequently, they can enjoy a much better scientific reputation. One may think that big intrinsical differences among researchers can constitute the main driving force behind these inequalities. Nevertheless, this model shows that, under specific circumstances, very similar individuals may self-organize themselves forming unequal and hierarchical structures.     

Computing the distribution of sequential Hölder norms of the Brownian motion

ABSTRACT

In this article, we give explicit formulas and study practical computations for the distribution function of sequential Hölder norms of a Brownian motion and of a Brownian bridge. We also discuss some statistical applications in the detection of some short “epidemic” changes in a sample.     

INFORMATION GATHERING IN ORGANIZATIONS: EQUILIBRIUM,WELFARE, AND OPTIMAL NETWORK STRUCTURE

We model an organization as a game in which all agents share a common decision problem and some level of coordination is necessary between individual actions. Agents have individual private information concerning the task they have to perform, and they share this private information through pairwise channels of communication. We analyze how this communication pattern, modeled by means of a network structure, affects individual behavior and aggregate welfare. In the unique equilibrium of this Bayesian game, each agent's optimal action depends on a properly defined knowledge index that measures how the aggregation of information helps him to infer higher‐order beliefs about other's information after communication. Adding communication channels is not always beneficial for the organization because it can lead to mis‐coordination. We single out the geometry of interagent communication links that the manager could implement in order to improve the organization's performance.     

Testing the Vitamin Model of job stress in Dutch health care workers

Three central hypotheses of Warr's Vitamin Model concerning the relationship between job characteristics and well-being and health outcomes were tested: (1) differential effects of job characteristics on the various well-being and health outcomes; (2) predominance of curvilinear associations; and (3) moderate influence of negative and positive affectivity on these relationships. The study participants were 162 employees from a health care organization (aged 19–54 years, 95% women) who completed questionnaires on job demands and job autonomy, as well as on the outcome variables depression, anxiety, job satisfaction, and health complaints. In addition, data on short-term sickness absence were collected. A higher level of job demands was significantly associated with a lower level of well-being and self-reported health. Job autonomy showed weaker relationships with the outcome variables. The effects of job demands were still large after controlling for negative and positive affectivity, while the effects of job autonomy in most cases became non-significant. The predicted curvilinear relationship between job characteristics and outcome variables did not have an additional value over a linear model in predicting the data. It is concluded that the present data from a homogeneous sample of mostly female nurses support Warr's Vitamin Model to a limited extent.     

Cost‐Benefit Analysis for Optimization of Risk Protection Under Budget Constraints

Cost‐benefit analysis (CBA) is commonly applied as a tool for deciding on risk protection. With CBA, one can identify risk mitigation strategies that lead to an optimal tradeoff between the costs of the mitigation measures and the achieved risk reduction. In practical applications of CBA, the strategies are typically evaluated through efficiency indicators such as the benefit‐cost ratio (BCR) and the marginal cost (MC) criterion. In many of these applications, the BCR is not consistently defined, which, as we demonstrate in this article, can lead to the identification of suboptimal solutions. This is of particular relevance when the overall budget for risk reduction measures is limited and an optimal allocation of resources among different subsystems is necessary. We show that this problem can be formulated as a hierarchical decision problem, where the general rules and decisions on the available budget are made at a central level (e.g., central government agency, top management), whereas the decisions on the specific measures are made at the subsystem level (e.g., local communities, company division). It is shown that the MC criterion provides optimal solutions in such hierarchical optimization. Since most practical applications only include a discrete set of possible risk protection measures, the MC criterion is extended to this situation. The findings are illustrated through a hypothetical numerical example. This study was prepared as part of our work on the optimal management of natural hazard risks, but its conclusions also apply to other fields of risk management.     

On vertex-parity edge-colorings

A vertex signature \(\pi \) of a finite graph G is any mapping \(\pi \,{:}\,V(G)\rightarrow \{0,1\}\). An edge-coloring of G is said to be vertex-parity for the pair \((G,\pi )\) if for every vertex v each color used on the edges incident to v appears in parity accordance with \(\pi \), i.e. an even or odd number of times depending on whether \(\pi (v)\) equals 0 or 1, respectively. The minimum number of colors for which \((G,\pi )\) admits such an edge-coloring is denoted by \(\chi '_p(G,\pi )\). We characterize the existence and prove that \(\chi '_p(G,\pi )\) is at most 6. Furthermore, we give a structural characterization of the pairs \((G,\pi )\) for which \(\chi '_p(G,\pi )=5\) and \(\chi '_p(G,\pi )=6\). In the last part of the paper, we consider a weaker version of the coloring, where it suffices that at every vertex, at least one color appears in parity accordance with \(\pi \). We show that the corresponding chromatic index is at most 3 and give a complete characterization for it.     

Reliable error estimation for Sobol’ indices

In the field of sensitivity analysis, Sobol’ indices are sensitivity measures widely used to assess the importance of inputs of a model to its output. The estimation of these indices is often performed through Monte Carlo or quasi-Monte Carlo methods. A notable method is the replication procedure that estimates first-order indices at a reduced cost in terms of number of model evaluations. An inherent practical problem of this estimation is how to quantify the number of model evaluations needed to ensure that estimates satisfy a desired error tolerance. This article addresses this challenge by proposing a reliable error bound for first-order and total effect Sobol’ indices. Starting from the integral formula of the indices, the error bound is defined in terms of the discrete Walsh coefficients of the different integrands. We propose a sequential estimation procedure of Sobol’ indices using the error bound as a stopping criterion. The sequential procedure combines Sobol’ sequences with either Saltelli’s strategy to estimate both first-order and total effect indices, or the replication procedure to estimate only first-order indices.