Beyond tail median and conditional tail expectation: Extreme risk estimation using tail Lp-optimization

The conditional tail expectation (CTE) is an indicator of tail behavior that takes into account both the frequency and magnitude of a tail event. However, the asymptotic normality of its empirical estimator requires that the underlying distribution possess a finite variance; this can be a strong restriction in actuarial and financial applications. A valuable alternative is the median shortfall (MS), although it only gives information about the frequency of a tail event. We construct a class of tail L^p-medians encompassing the MS and CTE. For p in (1,2), a tail L^p-median depends on both the frequency and magnitude of tail events, and its empirical estimator is, within the range of the data, asymptotically normal under a condition weaker than a finite variance. We extrapolate this estimator and another technique to extreme levels using the heavy-tailed framework. The estimators are showcased on a simulation study and on real fire insurance data.     

CORRUPTION: TOP DOWN OR BOTTOM UP?

This article studies the impact of corruption on an economy with a hierarchical government. In particular, we study whether centralizing corruption within the higher level of government increases or decreases the total amount of corruption. We show that when the after-tax relative profitability of the formal sector as compared to that of the informal sector is high enough, adding a layer of government increases the total amount of corruption. By contrast, for high-enough public wages and/or an efficient monitoring technology of the bureaucratic system, centralization of corruption at the top of the government hierarchy redistributes bribe income from the lower level to the upper level. In the process, total corruption is reduced and the formal sector of the economy expands.     

Conditional Likelihood Estimators for Hidden Markov Models and Stochastic Volatility Models

ABSTRACT.  This paper develops a new contrast process for parametric inference of general hidden Markov models, when the hidden chain has a non-compact state space. This contrast is based on the conditional likelihood approach, often used for ARCH-type models. We prove the strong consistency of the conditional likelihood estimators under appropriate conditions. The method is applied to the Kalman filter (for which this contrast and the exact likelihood lead to asymptotically equivalent estimators) and to the discretely observed stochastic volatility models.     

Economics of Radiation Protection: Equity Considerations

In order to implement cost-benefit analysis of protective actions to reduce radiological exposures, one needs to attribute a monetary value to the avoided exposure. Recently, the International Commission on Radiological Protection has stressed the need to take into consideration not only the collective exposure to ionising radiation but also its dispersion in the population. In this paper, by using some well known and some recent results in the economics of uncertainty, we discuss how to integrate these recommendations in the valuation of the benefit of protection.     

“Phil's Calling Grandma… ”: The Role of External Support in Human–Companion Animal Identity Pairings

This exploratory paper deals with human–animal role identity pairings such as parent–child or sibling–sibling and the necessity of support from other actors both for the formation of these idiosyncratic identities, as well as for their situational placement in social environments not limited to the nonhuman animal. Taken from a qualitative study examining identity formation counter to the nonhuman animal, I use in‐depth interviews of both people with and without human children to demonstrate how human‐to‐human relationships are formed by categorizing the companion animal as a “child” of sorts within the family structure. These relationships prove integral to the continued development and enactment of identities such as the animal “parent” or the animal “sibling” via three different groups: their own parents, partners, and, in one case, adult siblings. This creates positive affect and commitment to the identity across other social situations. Implications of these findings for identity theory and family research are discussed.     

Data-driven Kriging models based on FANOVA-decomposition

Kriging models have been widely used in computer experiments for the analysis of time-consuming computer codes. Based on kernels, they are flexible and can be tuned to many situations. In this paper, we construct kernels that reproduce the computer code complexity by mimicking its interaction structure. While the standard tensor-product kernel implicitly assumes that all interactions are active, the new kernels are suited for a general interaction structure, and will take advantage of the absence of interaction between some inputs. The methodology is twofold. First, the interaction structure is estimated from the data, using a first initial standard Kriging model, and represented by a so-called FANOVA graph. New FANOVA-based sensitivity indices are introduced to detect active interactions. Then this graph is used to derive the form of the kernel, and the corresponding Kriging model is estimated by maximum likelihood. The performance of the overall procedure is illustrated by several 3-dimensional and 6-dimensional simulated and real examples. A substantial improvement is observed when the computer code has a relatively high level of complexity.     

Cytometry inference through adaptive atomic deconvolution

In this paper, we consider a statistical estimation problem known as atomic deconvolution. Introduced in reliability, this model has a direct application when considering biological data produced by flow cytometers. From a statistical point of view, we aim at inferring the percentage of cells expressing the selected molecule and the probability distribution function associated with its fluorescence emission. We propose here an adaptive estimation procedure based on a previous deconvolution procedure introduced by Es, Gugushvili, and Spreij [(2008), ‘Deconvolution for an atomic distribution’, Electronic Journal of Statistics, 2, 265–297] and Gugushvili, Es, and Spreij [(2011), ‘Deconvolution for an atomic distribution: rates of convergence’, Journal of Nonparametric Statistics, 23, 1003–1029]. For both estimating the mixing parameter and the mixing density automatically, we use the Lepskii method based on the optimal choice of a bandwidth using a bias-variance decomposition. We then derive some convergence rates that are shown to be minimax optimal (up to some log terms) in Sobolev classes. Finally, we apply our algorithm on the simulated and real biological data.     

Early Segmentation Abilities in Preterm Infants

Infants born preterm have higher risks of developing linguistic deficits. Considering that the ability to segment words from fluent speech is crucial for lexical acquisition, Experiment 1 tested the ability of healthy extremely‐to‐late preterm infants to segment monosyllabic words at 6 months of postnatal age. Results establish basic segmentation skills in these infants. While we failed to find an effect of the degree of prematurity, this issue will need further exploration. Future studies will also have to specify the scope of these early segmentation skills, both in terms of the types of words segmented, the cues used to do so, and in terms of possible differences in performance between subgroups of preterm infants (e.g., based on gestational age or medical risks). Lastly, given that the preterm infants tested had a mean maturational age of 4 months, Experiment 2 explored monosyllabic segmentation in full‐term 4‐month olds. Infants succeeded at the task, providing the earliest developmental evidence of word segmentation in full‐term infants. These findings better specify the early trajectory of segmentation abilities in both full‐term and healthy, low‐risk preterm infants and support the proposal that prematurity might have a differential effect on the early acquisition of various linguistic levels.     

Food in a row: urban trees offer valuable floral resources to pollinating insects

Urbanization affects the availability and diversity of floral resources (pollen and/or nectar) for wild pollinating insects. For example, urban green areas are characterized by an abundance of ornamental plant species. Increasingly, trees are planted to improve the aesthetics of urban streets and parks. These urban trees might offer important floral resources to pollinating insects. To examine the suitability of urban trees as resources for pollinating insects, we investigated the chemical composition of pollen and nectar as well as the amount of nectar produced by the nine major insect-pollinated tree species planted in cities of Western Europe, namely Acer pseudoplatanus, Aesculus carnea, A. hippocastanum, Robinia pseudoacacia, Tilia cordata, T. x euchlora, T. x europaea, T. platyphyllos and T. tomentosa. The analyses revealed that globally the Tilia trees provide pollen with lower contents of polypeptides, amino acids and phytosterols compared with the other species. Urban tree flowers offer abundant nectar with relatively high sugar contents (0.16–1.28 mg/flower); sucrose was the predominant sugar in all nectars. The investigated tree species could therefore be considered in future city plantings.