Rényi entropy of m-generalized order statistics

Abstract

Characterizing relations via Rényi entropy of m-generalized order statistics are considered along with examples and related stochastic orderings. Previous results for common order statistics are included.     

Integrating Operational and Organizational Aspects in Interdependent Infrastructure Network Recovery

Managing risk in infrastructure systems implies dealing with interdependent physical networks and their relationships with the natural and societal contexts. Computational tools are often used to support operational decisions aimed at improving resilience, whereas economics‐related tools tend to be used to address broader societal and policy issues in infrastructure management. We propose an optimization‐based framework for infrastructure resilience analysis that incorporates organizational and socioeconomic aspects into operational problems, allowing to understand relationships between decisions at the policy level (e.g., regulation) and the technical level (e.g., optimal infrastructure restoration). We focus on three issues that arise when integrating such levels. First, optimal restoration strategies driven by financial and operational factors evolve differently compared to those driven by socioeconomic and humanitarian factors. Second, regulatory aspects have a significant impact on recovery dynamics (e.g., effective recovery is most challenging in societies with weak institutions and regulation, where individual interests may compromise societal well‐being). And third, the decision space (i.e., available actions) in postdisaster phases is strongly determined by predisaster decisions (e.g., resource allocation). The proposed optimization framework addresses these issues by using: (1) parametric analyses to test the influence of operational and socioeconomic factors on optimization outcomes, (2) regulatory constraints to model and assess the cost and benefit (for a variety of actors) of enforcing specific policy‐related conditions for the recovery process, and (3) sensitivity analyses to capture the effect of predisaster decisions on recovery. We illustrate our methodology with an example regarding the recovery of interdependent water, power, and gas networks in Shelby County, TN (USA), with exposure to natural hazards.     

Risk Modeling of Interdependent Complex Systems of Systems: Theory and Practice

The emergence of the complexity characterizing our systems of systems (SoS) requires a reevaluation of the way we model, assess, manage, communicate, and analyze the risk thereto. Current models for risk analysis of emergent complex SoS are insufficient because too often they rely on the same risk functions and models used for single systems. These models commonly fail to incorporate the complexity derived from the networks of interdependencies and interconnectedness (I–I) characterizing SoS. There is a need to reevaluate currently practiced risk analysis to respond to this reality by examining, and thus comprehending, what makes emergent SoS complex. The key to evaluating the risk to SoS lies in understanding the genesis of characterizing I–I of systems manifested through shared states and other essential entities within and among the systems that constitute SoS. The term “essential entities” includes shared decisions, resources, functions, policies, decisionmakers, stakeholders, organizational setups, and others. This undertaking can be accomplished by building on state‐space theory, which is fundamental to systems engineering and process control. This article presents a theoretical and analytical framework for modeling the risk to SoS with two case studies performed with the MITRE Corporation and demonstrates the pivotal contributions made by shared states and other essential entities to modeling and analysis of the risk to complex SoS. A third case study highlights the multifarious representations of SoS, which require harmonizing the risk analysis process currently applied to single systems when applied to complex SoS.     

Decision‐making under uncertainty in child protection: Creating a just and learning culture

The argument is made for having a positive error culture in child protection to improve decision‐making and risk management. This requires organizations to accept that mistakes are likely and to treat them as opportunities for learning and improving. In contrast, in many organizations, a punitive reaction to errors leads to workers hiding them and developing a defensive approach to their practice with children and families. The safety management literature has shown how human error is generally not simply due to a “bad apple” but made more or less likely by the work context that helps or hinders good performance. Improving safety requires learning about the weaknesses in the organization that contribute to poor performance. To create a learning culture, people need to feel that when they talk about mistakes or weak practice, there will be a constructive response from their organization. One aspect of reducing the blame culture is to develop a shared understanding of how practice will be judged and how those appraising practice will avoid the hindsight bias. To facilitate a positive error culture, a set of risk principles are presented that offer a set of criteria by which practice should be appraised.     

On the Convergence of the Monte Carlo Maximum Likelihood Method for Latent Variable Models

While much used in practice, latent variable models raise challenging estimation problems due to the intractability of their likelihood. Monte Carlo maximum likelihood (MCML), as proposed by Geyer & Thompson (1992 ), is a simulation-based approach to maximum likelihood approximation applicable to general latent variable models. MCML can be described as an importance sampling method in which the likelihood ratio is approximated by Monte Carlo averages of importance ratios simulated from the complete data model corresponding to an arbitrary value of the unknown parameter. This paper studies the asymptotic (in the number of observations) performance of the MCML method in the case of latent variable models with independent observations. This is in contrast with previous works on the same topic which only considered conditional convergence to the maximum likelihood estimator, for a fixed set of observations. A first important result is that when is fixed, the MCML method can only be consistent if the number of simulations grows exponentially fast with the number of observations. If on the other hand, is obtained from a consistent sequence of estimates of the unknown parameter, then the requirements on the number of simulations are shown to be much weaker.     

Spatial variation in soil inorganic nitrogen across an arid urban ecosystem

We explored variations in inorganic soil nitrogen (N) concentrations across metropolitan Phoenix, Arizona, and the surrounding desert using a probability-based synoptic survey. Data were examined using spatial statistics on the entire region, as well as for the desert and urban sites separately. Concentrations of both NO₃-N and NH₄-N were markedly higher and more heterogeneous amongst urban compared to desert soils. Regional variation in soil NO₃-N concentration was best explained by latitude, land use history, population density, along with percent cover of impervious surfaces and lawn, whereas soil NH₄-N concentrations were related to only latitude and population density. Within the urban area, patterns in both soil NO₃-N and NH₄-N were best predicted by elevation, population density and type of irrigation in the surrounding neighborhood. Spatial autocorrelation of soil NO₃-N concentrations explained 49% of variation among desert sites but was absent between urban sites. We suggest that inorganic soil N concentrations are controlled by a number of ‘local’ or ‘neighborhood’ human-related drivers in the city, rather than factors related to an urban-rural gradient.     

Challenge of multi-level governance in developing countries and possible GIS applications

Of the three most prominent issues in governance discourse—the horizontal coordination of markets, government hierarchies and networks, global governance and multi-level governance—this paper argues that the last, developing and sustaining structures of multi-level governance, constitutes the most profound governance challenge for most developing countries.The paper identifies the major forces promoting change in favor of multi-level governance in developing countries. Using the growing literature and comparative national experiences (illustrated with the examples of India, the Philippines, Colombia and Nigeria), it highlights the key achievements of democratic decentralization as well as its nagging problems in these countries. Finally, the paper demonstrates the significance of these developments for democracy and development in these countries and shows some of the ways by which external partners—especially experts associated with the Geographic Information Systems for Developing Countries can assist and possibly benefit from this process.     

Statistical and Combinatorial Aspects of Comparative Genomics*

Abstract. This document presents a survey of the statistical and combinatorial aspects of four areas of comparative genomics: gene order based measures of evolutionary distances between species, construction of phylogenetic trees, detection of horizontal transfer of genes, and detection of ancient whole genome duplications.     

'Polyphonic' welfare: Luhmann's systems theory applied to modern social work

This article applies a series of concepts from Niklas Luhmann's systems theory in an analysis of modern welfare organisations. The point of departure is that social help in late modern welfare states has become 'polycentric' in that 'help' is today being defined by various different agents: public, voluntary and private care providers. Empirically, this article investigates re-housing work with homeless people, a kind of social work which involves several different welfare organisations. The case study shows how these organisations define themselves by making internal constructions of their surroundings, and how their self-enclosed nature creates a certain 'insensitivity' towards one another. How to coordinate and translate within this 'polyphony' of incomparable observations and values represents a major managerial challenge for present-day social workers.     

Fast retraction of evidence in a probabilistic expert system

We present and justify a propagation algorithm to facilitate the simultaneous calculation, for every node in a probabilistic exper system of the distribution of the associated random quantity, conditional on all the evidence obtained about the remaining nodes.