Flood Protection Diversification to Reduce Probabilities of Extreme Losses

Recent catastrophic losses because of floods require developing resilient approaches to flood risk protection. This article assesses how diversification of a system of coastal protections might decrease the probabilities of extreme flood losses. The study compares the performance of portfolios each consisting of four types of flood protection assets in a large region of dike rings. A parametric analysis suggests conditions in which diversifications of the types of included flood protection assets decrease extreme flood losses. Increased return periods of extreme losses are associated with portfolios where the asset types have low correlations of economic risk. The effort highlights the importance of understanding correlations across asset types in planning for large‐scale flood protection. It allows explicit integration of climate change scenarios in developing flood mitigation strategy.     

Remittance Activity among Brazilians in the US and Canada

This comparative study uses data collected in 1990 and 1991 to examine the remittance behavior of Brazilians who had recently arrived in Canada and the United States. These data permit an examination of remittance activity among immigrants relatively soon after their arrival in a pair of host destinations. Prior to contrasting the remittance activity of these newly arrived immigrants, we first document the high degree of similarity between the two groups at their time of arrival; a point that becomes important when contrasting their divergent outcomes. Next, this study contributes to the research literature on micro‐level remittance patterns and behaviors by focusing on three policy relevant dependant variables. More specifically, multivariate analyses are undertaken to examine those individual‐level factors that best determine: (1) who remits, (2) how much they remit, and (3) when funds are remitted for productive purposes. Because the data analyzed were collected with the same instrument, results are then contrasted for the two destinations. One major finding to emerge from this cross‐national study was that even immigrants who are extremely similar on all socio‐economic measures at time of arrival may soon manifest divergent outcomes due to their distinctive country of destination experiences. Consequently, although some common predictors were found in both locales, there were far fewer than expected. For example, family obligation variables were usually significant in the predicted directions for the United States data, while this was almost never the case for the Canadian data. Furthermore, some variables often assumed to predict remittance behavior were insignificant in both locations. This study concludes by considering possible explanations for these results, as well as discussing the need for additional theoretical work and data collection in the area of immigrant remittance activity.     

Poisson-Driven Stationary Markov Models

ABSTRACT

We propose a simple yet powerful method to construct strictly stationary Markovian models with given but arbitrary invariant distributions. The idea is based on a Poisson-type transform modulating the dependence structure in the model. An appealing feature of our approach is the possibility to control the underlying transition probabilities and, therefore, incorporate them within standard estimation methods. Given the resulting representation of the transition density, a Gibbs sampler algorithm based on the slice method is proposed and implemented. In the discrete-time case, special attention is placed to the class of generalized inverse Gaussian distributions. In the continuous case, we first provide a brief treatment of the class of gamma distributions, and then extend it to cover other invariant distributions, such as the generalized extreme value class. The proposed approach and estimation algorithm are illustrated with real financial datasets. Supplementary materials for this article are available online.     

Resilience Analytics with Application to Power Grid of a Developing Region

Infrastructure development of volatile regions is a significant investment by international government and nongovernment organizations, with attendant requirements for risk management. Global development banks may be tasked to manage these investments and provide a channel between donors and borrowers. Moreover, various stakeholders from the private sector, local and international agencies, and the military can be engaged in conception, planning, and implementation of constituent projects. Emergent and future conditions of military conflict, politics, economics, technology, environment, behaviors, institutions, and society that stress infrastructure development are prevalent, and funding mechanisms are vulnerable to fraud, waste, and abuse. This article will apply resilience analytics with scenario‐based preferences to identify the stressors that most influence a prioritization of initiatives in the electric power sector of Afghanistan. The resilience in this article is conceived in terms of the degree of disruption of priorities when stressors influence the preferences of stakeholders, and ultimately a prioritization of initiatives. The ancillary results include an understanding of which initiatives contribute most and least across strategic criteria and which criteria have the most impact for the analysis. The article concludes with recommendations for risk monitoring and risk management of the portfolio of stressors through the life cycle and horizon of grid capacity expansion.     

Conditions equivalent and doubly equivalent to consistency of approximate MLE's

Summary Strasser (1981) introduced approximately maximum likelihood estimators (AMLE's) and found a condition equivalent to strong consistency of all AMLE's. Here a condition weaker than that of Strasser is proved to be equivalent to the usual consistency of all AMLE's. Under an additional regularity this condition is shown to be doubly equivalent, which means that it is equivalent to consistency, and its contrary is equivalent to inconsistency of all AMLE's. The doubly equivalent conditions are important—we present an example where MLE is strongly consistent but some AMLE's are inconsistent. It is proved that the additional regularity can be reduced to the finiteness of an observations entropy. All results are motivate and illustrated by examples. Supported by CSAS grant N. 17503.     

Sampling hierarchies of discrete random structures

Statistics and Computing - Hierarchical normalized discrete random measures identify a general class of priors that is suited to flexibly learn how the distribution of a response variable changes...     

Flexible clustering via hidden hierarchical Dirichlet priors

The Bayesian approach to inference stands out for naturally allowing borrowing information across heterogeneous populations, with different samples possibly sharing the same distribution. A popular Bayesian nonparametric model for clustering probability distributions is the nested Dirichlet process, which however has the drawback of grouping distributions in a single cluster when ties are observed across samples. With the goal of achieving a flexible and effective clustering method for both samples and observations, we investigate a nonparametric prior that arises as the composition of two different discrete random structures and derive a closed-form expression for the induced distribution of the random partition, the fundamental tool regulating the clustering behavior of the model. On the one hand, this allows to gain a deeper insight into the theoretical properties of the model and, on the other hand, it yields an MCMC algorithm for evaluating Bayesian inferences of interest. Moreover, we single out limitations of this algorithm when working with more than two populations and, consequently, devise an alternative more efficient sampling scheme, which as a by-product, allows testing homogeneity between different populations. Finally, we perform a comparison with the nested Dirichlet process and provide illustrative examples of both synthetic and real data.