Extremal properties of certain risk models

Cette thèse s'intéresse à étudier les propriétés extrémales de certains modèles de risque d'intérêt dans diverses applications de l'assurance, de la finance et des statistiques. Cette thèse se développe selon deux axes principaux, à savoir: Dans la première partie, nous nous concentrons sur deux modèles de risques univariés, c'est-à- dire, un modèle de risque de déflation et un modèle de risque de réassurance. Nous étudions le développement des queues de distribution sous certaines conditions des risques commun¬s. Les principaux résultats sont ainsi illustrés par des exemples typiques et des simulations numériques. Enfin, les résultats sont appliqués aux domaines des assurances, par exemple, les approximations de Value-at-Risk, d'espérance conditionnelle unilatérale etc. La deuxième partie de cette thèse est consacrée à trois modèles à deux variables: Le premier modèle concerne la censure à deux variables des événements extrême. Pour ce modèle, nous proposons tout d'abord une classe d'estimateurs pour les coefficients de dépendance et la probabilité des queues de distributions. Ces estimateurs sont flexibles en raison d'un paramètre de réglage. Leurs distributions asymptotiques sont obtenues sous certaines condi¬tions lentes bivariées de second ordre. Ensuite, nous donnons quelques exemples et présentons une petite étude de simulations de Monte Carlo, suivie par une application sur un ensemble de données réelles d'assurance. L'objectif de notre deuxième modèle de risque à deux variables est l'étude de coefficients de dépendance des queues de distributions obliques et asymétriques à deux variables. Ces distri¬butions obliques et asymétriques sont largement utiles dans les applications statistiques. Elles sont générées principalement par le mélange moyenne-variance de lois normales et le mélange de lois normales asymétriques d'échelles, qui distinguent la structure de dépendance de queue comme indiqué par nos principaux résultats. Le troisième modèle de risque à deux variables concerne le rapprochement des maxima de séries triangulaires elliptiques obliques. Les résultats théoriques sont fondés sur certaines hypothèses concernant le périmètre aléatoire sous-jacent des queues de distributions. -- This thesis aims to investigate the extremal properties of certain risk models of interest in vari¬ous applications from insurance, finance and statistics. This thesis develops along two principal lines, namely: In the first part, we focus on two univariate risk models, i.e., deflated risk and reinsurance risk models. Therein we investigate their tail expansions under certain tail conditions of the common risks. Our main results are illustrated by some typical examples and numerical simu¬lations as well. Finally, the findings are formulated into some applications in insurance fields, for instance, the approximations of Value-at-Risk, conditional tail expectations etc. The second part of this thesis is devoted to the following three bivariate models: The first model is concerned with bivariate censoring of extreme events. For this model, we first propose a class of estimators for both tail dependence coefficient and tail probability. These estimators are flexible due to a tuning parameter and their asymptotic distributions are obtained under some second order bivariate slowly varying conditions of the model. Then, we give some examples and present a small Monte Carlo simulation study followed by an application on a real-data set from insurance. The objective of our second bivariate risk model is the investigation of tail dependence coefficient of bivariate skew slash distributions. Such skew slash distributions are extensively useful in statistical applications and they are generated mainly by normal mean-variance mixture and scaled skew-normal mixture, which distinguish the tail dependence structure as shown by our principle results. The third bivariate risk model is concerned with the approximation of the component-wise maxima of skew elliptical triangular arrays. The theoretical results are based on certain tail assumptions on the underlying random radius.

Detection of ingested cocaine-filled packets: comparison of filtered back projection CT with adaptive statistical iterative reconstructed images

Purpose: To evaluate the diagnostic value and image quality of CT with filtered back projection (FBP) compared with adaptive statistical iterative reconstructed images (ASIR) in body stuffers with ingested cocaine-filled packets.Methods and Materials: Twenty-nine body stuffers (mean age 31.9 years, 3 women) suspected for ingestion of cocaine-filled packets underwent routine-dose 64-row multidetector CT with FBP (120kV, pitch 1.375, 100-300 mA and automatic tube current modulation (auto mA), rotation time 0.7sec, collimation 2.5mm), secondarily reconstructed with 30 % and 60 % ASIR. In 13 (44.83%) out of the body stuffers cocaine-filled packets were detected, confirmed by exact analysis of the faecal content including verification of the number (range 1-25). Three radiologists independently and blindly evaluated anonymous CT examinations (29 FBP-CT and 68 ASIR-CT) for the presence and number of cocaine-filled packets indicating observers' confidence, and graded them for diagnostic quality, image noise, and sharpness. Sensitivity, specificity, area under the receiver operating curve (ROC) Az and interobserver agreement between the 3 radiologists for FBP-CT and ASIR-CT were calculated.Results: The increase of the percentage of ASIR significantly diminished the objective image noise (p<0.001). Overall sensitivity and specificity for the detection of the cocaine-filled packets were 87.72% and 76.15%, respectively. The difference of ROC area Az between the different reconstruction techniques was significant (p= 0.0101), that is 0.938 for FBP-CT, 0.916 for 30 % ASIR-CT, and 0.894 for 60 % ASIR-CT.Conclusion: Despite the evident image noise reduction obtained by ASIR, the diagnostic value for detecting cocaine-filled packets decreases, depending on the applied ASIR percentage.

Methods for testing association between uncertain genotypes and quantitative traits.

Interpretability and power of genome-wide association studies can be increased by imputing unobserved genotypes, using a reference panel of individuals genotyped at higher marker density. For many markers, genotypes cannot be imputed with complete certainty, and the uncertainty needs to be taken into account when testing for association with a given phenotype. In this paper, we compare currently available methods for testing association between uncertain genotypes and quantitative traits. We show that some previously described methods offer poor control of the false-positive rate (FPR), and that satisfactory performance of these methods is obtained only by using ad hoc filtering rules or by using a harsh transformation of the trait under study. We propose new methods that are based on exact maximum likelihood estimation and use a mixture model to accommodate nonnormal trait distributions when necessary. The new methods adequately control the FPR and also have equal or better power compared to all previously described methods. We provide a fast software implementation of all the methods studied here; our new method requires computation time of less than one computer-day for a typical genome-wide scan, with 2.5 M single nucleotide polymorphisms and 5000 individuals.

Is the ability of biocontrol fluorescent pseudomonads to produce the antifungal metabolite 2,4-diacetylphloroglucinol really synonymous with higher plant protection?

The antifungal compound 2,4-diacetylphloroglucinol (Phl) contributes to biocontrol in pseudomonads, but whether or not Phl(+) biocontrol pseudomonads display higher plant-protecting activity than Phl(-) biocontrol pseudomonads remains to be demonstrated. This issue was addressed by assessing 230 biocontrol fluorescent pseudomonads selected from a collection of 3132 bacterial isolates obtained from 63 soils worldwide. One-third of the biocontrol pseudomonads were Phl(+) and almost all Phl(+) isolates also produced hydrogen cyanide (HCN). The only Phl(+) HCN(-) strain did harbor hcn genes, but with the deletion of a 134 bp hcnC fragment corresponding to an ADP-binding motif. Statistical analysis of biocontrol isolate distributions indicated that Phl production ability was associated with superior disease suppression activity in the Pythium-cucumber and Fusarium-tomato pathosystems, but this was also the case with HCN production ability. However, HCN significance was not as strong, as indicated both by the comparison of Phl(-) HCN(+) and Phl(-) HCN(-) strains and by correlation analyses. This is the first population-level demonstration of the higher plant-protecting activity of Phl(+) biocontrol pseudomonads in comparison with Phl(-) biocontrol pseudomonads.

Predicting spatial patterns of plant species richness: a comparison of direct macroecological and species stacking approaches

Aim This study compares the direct, macroecological approach (MEM) for modelling species richness (SR) with the more recent approach of stacking predictions from individual species distributions (S-SDM). We implemented both approaches on the same dataset and discuss their respective theoretical assumptions, strengths and drawbacks. We also tested how both approaches performed in reproducing observed patterns of SR along an elevational gradient.Location Two study areas in the Alps of Switzerland.Methods We implemented MEM by relating the species counts to environmental predictors with statistical models, assuming a Poisson distribution. S-SDM was implemented by modelling each species distribution individually and then stacking the obtained prediction maps in three different ways - summing binary predictions, summing random draws of binomial trials and summing predicted probabilities - to obtain a final species count.Results The direct MEM approach yields nearly unbiased predictions centred around the observed mean values, but with a lower correlation between predictions and observations, than that achieved by the S-SDM approaches. This method also cannot provide any information on species identity and, thus, community composition. It does, however, accurately reproduce the hump-shaped pattern of SR observed along the elevational gradient. The S-SDM approach summing binary maps can predict individual species and thus communities, but tends to overpredict SR. The two other S-SDM approaches the summed binomial trials based on predicted probabilities and summed predicted probabilities - do not overpredict richness, but they predict many competing end points of assembly or they lose the individual species predictions, respectively. Furthermore, all S-SDM approaches fail to appropriately reproduce the observed hump-shaped patterns of SR along the elevational gradient.Main conclusions Macroecological approach and S-SDM have complementary strengths. We suggest that both could be used in combination to obtain better SR predictions by following the suggestion of constraining S-SDM by MEM predictions.

Modern statistical models for forensic fingerprint examinations : a critical review

Over the last decade, the development of statistical models in support of forensic fingerprint identification has been the subject of increasing research attention, spurned on recently by commentators who claim that the scientific basis for fingerprint identification has not been adequately demonstrated. Such models are increasingly seen as useful tools in support of the fingerprint identification process within or in addition to the ACE-V framework. This paper provides a critical review of recent statistical models from both a practical and theoretical perspective. This includes analysis of models of two different methodologies: Probability of Random Correspondence (PRC) models that focus on calculating probabilities of the occurrence of fingerprint configurations for a given population, and Likelihood Ratio (LR) models which use analysis of corresponding features of fingerprints to derive a likelihood value representing the evidential weighting for a potential source.

Suitability, success and sinks: how do predictions of nesting distributions relate to fitness parameters in high arctic waders?

AimAlthough habitat suitability maps derived from species distribution models (SDMs) are often assumed to highlight locations that can sustain healthy populations over time, the relationship between suitability scores and fitness parameters has rarely been tested thoroughly. LocationZackenberg Valley, north-east Greenland. MethodsUsing 14years of data (1997-2010) representing three wader species (dunlin Calidris alpina, sanderling Calidris alba and ruddy turnstone Arenaria interpres), we tested the relationships between modelled suitability and fitness parameters at nesting locations. ResultsAmong the three species examined, only the ruddy turnstone exhibited significant relationships between suitability and nest success, but over time rather than space. During years with extensive snow cover in the landscape, the nesting sites of ruddy turnstone occurred in different habitats than were typically used across years. Moreover, in years with extensive snow cover, the ruddy turnstone initiated nests later and suffered from higher egg predation rates. Main conclusionOur results suggest that SDMs derived from species occurrences that include years of low reproductive success may over-estimate the potential suitable habitat in the landscape. Whenever possible, variation in reproductive success should be considered when building models to inform species' response to environmental change. species' response to environmental change.

Use of DNA profiles for investigation using a simulated national DNA database: Part II. Statistical and ethical considerations on familial searching.

Familial searching consists of searching for a full profile left at a crime scene in a National DNA Database (NDNAD). In this paper we are interested in the circumstance where no full match is returned, but a partial match is found between a database member's profile and the crime stain. Because close relatives share more of their DNA than unrelated persons, this partial match may indicate that the crime stain was left by a close relative of the person with whom the partial match was found. This approach has successfully solved important crimes in the UK and the USA. In a previous paper, a model, which takes into account substructure and siblings, was used to simulate a NDNAD. In this paper, we have used this model to test the usefulness of familial searching and offer guidelines for pre-assessment of the cases based on the likelihood ratio. Siblings of "persons" present in the simulated Swiss NDNAD were created. These profiles (N=10,000) were used as traces and were then compared to the whole database (N=100,000). The statistical results obtained show that the technique has great potential confirming the findings of previous studies. However, effectiveness of the technique is only one part of the story. Familial searching has juridical and ethical aspects that should not be ignored. In Switzerland for example, there are no specific guidelines to the legality or otherwise of familial searching. This article both presents statistical results, and addresses criminological and civil liberties aspects to take into account risks and benefits of familial searching.

JULIDE: a software tool for 3D reconstruction and statistical analysis of autoradiographic mouse brain sections.

In this article we introduce JULIDE, a software toolkit developed to perform the 3D reconstruction, intensity normalization, volume standardization by 3D image registration and voxel-wise statistical analysis of autoradiographs of mouse brain sections. This software tool has been developed in the open-source ITK software framework and is freely available under a GPL license. The article presents the complete image processing chain from raw data acquisition to 3D statistical group analysis. Results of the group comparison in the context of a study on spatial learning are shown as an illustration of the data that can be obtained with this tool.

Statistical discrimination of steroid profiles in doping control with support vector machines.

Due to their performance enhancing properties, use of anabolic steroids (e.g. testosterone, nandrolone, etc.) is banned in elite sports. Therefore, doping control laboratories accredited by the World Anti-Doping Agency (WADA) screen among others for these prohibited substances in urine. It is particularly challenging to detect misuse with naturally occurring anabolic steroids such as testosterone (T), which is a popular ergogenic agent in sports and society. To screen for misuse with these compounds, drug testing laboratories monitor the urinary concentrations of endogenous steroid metabolites and their ratios, which constitute the steroid profile and compare them with reference ranges to detect unnaturally high values. However, the interpretation of the steroid profile is difficult due to large inter-individual variances, various confounding factors and different endogenous steroids marketed that influence the steroid profile in various ways. A support vector machine (SVM) algorithm was developed to statistically evaluate urinary steroid profiles composed of an extended range of steroid profile metabolites. This model makes the interpretation of the analytical data in the quest for deviating steroid profiles feasible and shows its versatility towards different kinds of misused endogenous steroids. The SVM model outperforms the current biomarkers with respect to detection sensitivity and accuracy, particularly when it is coupled to individual data as stored in the Athlete Biological Passport.

A temporal hierarchy for conspecific vocalization discrimination in humans.

Introduction: Discrimination of species-specific vocalizations is fundamental for survival and social interactions. Its unique behavioral relevance has encouraged the identification of circumscribed brain regions exhibiting selective responses (Belin et al., 2004), while the role of network dynamics has received less attention. Those studies that have examined the brain dynamics of vocalization discrimination leave unresolved the timing and the inter-relationship between general categorization, attention, and speech-related processes (Levy et al., 2001, 2003; Charest et al., 2009). Given these discrepancies and the presence of several confounding factors, electrical neuroimaging analyses were applied to auditory evoked-potential (AEPs) to acoustically and psychophysically controlled non-verbal human and animal vocalizations. This revealed which region(s) exhibit voice-sensitive responses and in which sequence. Methods: Subjects (N=10) performed a living vs. man-made 'oddball' auditory discrimination task, such that on a given block of trials 'target' stimuli occurred 10% of the time. Stimuli were complex, meaningful sounds of 500ms duration. There were 120 different sound files in total, 60 of which represented sounds of living objects and 60 man-made objects. The stimuli that were the focus of the present investigation were restricted to those of living objects within blocks where no response was required. These stimuli were further sorted between human non-verbal vocalizations and animal vocalizations. They were also controlled in terms of their spectrograms and formant distributions. Continuous 64-channel EEG was acquired through Neuroscan Synamps referenced to the nose, band-pass filtered 0.05-200Hz, and digitized at 1000Hz. Peri-stimulus epochs of continuous EEG (-100ms to 900ms) were visually inspected for artifacts, 40Hz low-passed filtered and baseline corrected using the pre-stimulus period . Averages were computed from each subject separately. AEPs in response to animal and human vocalizations were analyzed with respect to differences of Global Field Power (GFP) and with respect to changes of the voltage configurations at the scalp (reviewed in Murray et al., 2008). The former provides a measure of the strength of the electric field irrespective of topographic differences; the latter identifies changes in spatial configurations of the underlying sources independently of the response strength. In addition, we utilized the local auto-regressive average distributed linear inverse solution (LAURA; Grave de Peralta Menendez et al., 2001) to visualize and statistically contrast the likely underlying sources of effects identified in the preceding analysis steps. Results: We found differential activity in response to human vocalizations over three periods in the post-stimulus interval, and this response was always stronger than that to animal vocalizations. The first differential response (169-219ms) was a consequence of a modulation in strength of a common brain network localized into the right superior temporal sulcus (STS; Brodmann's Area (BA) 22) and extending into the superior temporal gyrus (STG; BA 41). A second difference (291-357ms) also followed from strength modulations of a common network with statistical differences localized to the left inferior precentral and prefrontal gyrus (BA 6/45). These two first strength modulations correlated (Spearman's rho(8)=0.770; p=0.009) indicative of functional coupling between temporally segregated stages of vocalization discrimination. A third difference (389-667ms) followed from strength and topographic modulations and was localized to the left superior frontal gyrus (BA10) although this third difference did not reach our spatial criterion of 12 continuous voxels. Conclusions: We show that voice discrimination unfolds over multiple temporal stages, involving a wide network of brain regions. The initial stages of vocalization discrimination are based on modulations in response strength within a common brain network with no evidence for a voice-selective module. The latency of this effect parallels that of face discrimination (Bentin et al., 2007), supporting the possibility that voice and face processes can mutually inform one another. Putative underlying sources (localized in the right STS; BA 22) are consistent with prior hemodynamic imaging evidence in humans (Belin et al., 2004). Our effect over the 291-357ms post-stimulus period overlaps the 'voice-specific-response' reported by Levy et al. (Levy et al., 2001) and the estimated underlying sources (left BA6/45) were in agreement with previous findings in humans (Fecteau et al., 2005). These results challenge the idea that circumscribed and selective areas subserve con-specific vocalization processing.