On optimal dividend strategies with deficit or incomplete information

Rapport de synthèse Cette thèse consiste en trois essais sur les stratégies optimales de dividendes. Chaque essai correspond à un chapitre. Les deux premiers essais ont été écrits en collaboration avec les Professeurs Hans Ulrich Gerber et Elias S. W. Shiu et ils ont été publiés; voir Gerber et al. (2006b) ainsi que Gerber et al. (2008). Le troisième essai a été écrit en collaboration avec le Professeur Hans Ulrich Gerber. Le problème des stratégies optimales de dividendes remonte à de Finetti (1957). Il se pose comme suit: considérant le surplus d'une société, déterminer la stratégie optimale de distribution des dividendes. Le critère utilisé consiste à maximiser la somme des dividendes escomptés versés aux actionnaires jusqu'à la ruine2 de la société. Depuis de Finetti (1957), le problème a pris plusieurs formes et a été résolu pour différents modèles. Dans le modèle classique de théorie de la ruine, le problème a été résolu par Gerber (1969) et plus récemment, en utilisant une autre approche, par Azcue and Muler (2005) ou Schmidli (2008). Dans le modèle classique, il y a un flux continu et constant d'entrées d'argent. Quant aux sorties d'argent, elles sont aléatoires. Elles suivent un processus à sauts, à savoir un processus de Poisson composé. Un exemple qui correspond bien à un tel modèle est la valeur du surplus d'une compagnie d'assurance pour lequel les entrées et les sorties sont respectivement les primes et les sinistres. Le premier graphique de la Figure 1 en illustre un exemple. Dans cette thèse, seules les stratégies de barrière sont considérées, c'est-à-dire quand le surplus dépasse le niveau b de la barrière, l'excédent est distribué aux actionnaires comme dividendes. Le deuxième graphique de la Figure 1 montre le même exemple du surplus quand une barrière de niveau b est introduite, et le troisième graphique de cette figure montre, quand à lui, les dividendes cumulés. Chapitre l: "Maximizing dividends without bankruptcy" Dans ce premier essai, les barrières optimales sont calculées pour différentes distributions du montant des sinistres selon deux critères: I) La barrière optimale est calculée en utilisant le critère usuel qui consiste à maximiser l'espérance des dividendes escomptés jusqu'à la ruine. II) La barrière optimale est calculée en utilisant le second critère qui consiste, quant à lui, à maximiser l'espérance de la différence entre les dividendes escomptés jusqu'à la ruine et le déficit au moment de la ruine. Cet essai est inspiré par Dickson and Waters (2004), dont l'idée est de faire supporter aux actionnaires le déficit au moment de la ruine. Ceci est d'autant plus vrai dans le cas d'une compagnie d'assurance dont la ruine doit être évitée. Dans l'exemple de la Figure 1, le déficit au moment de la ruine est noté R. Des exemples numériques nous permettent de comparer le niveau des barrières optimales dans les situations I et II. Cette idée, d'ajouter une pénalité au moment de la ruine, a été généralisée dans Gerber et al. (2006a). Chapitre 2: "Methods for estimating the optimal dividend barrier and the probability of ruin" Dans ce second essai, du fait qu'en pratique on n'a jamais toute l'information nécessaire sur la distribution du montant des sinistres, on suppose que seuls les premiers moments de cette fonction sont connus. Cet essai développe et examine des méthodes qui permettent d'approximer, dans cette situation, le niveau de la barrière optimale, selon le critère usuel (cas I ci-dessus). Les approximations "de Vylder" et "diffusion" sont expliquées et examinées: Certaines de ces approximations utilisent deux, trois ou quatre des premiers moments. Des exemples numériques nous permettent de comparer les approximations du niveau de la barrière optimale, non seulement avec les valeurs exactes mais également entre elles. Chapitre 3: "Optimal dividends with incomplete information" Dans ce troisième et dernier essai, on s'intéresse à nouveau aux méthodes d'approximation du niveau de la barrière optimale quand seuls les premiers moments de la distribution du montant des sauts sont connus. Cette fois, on considère le modèle dual. Comme pour le modèle classique, dans un sens il y a un flux continu et dans l'autre un processus à sauts. A l'inverse du modèle classique, les gains suivent un processus de Poisson composé et les pertes sont constantes et continues; voir la Figure 2. Un tel modèle conviendrait pour une caisse de pension ou une société qui se spécialise dans les découvertes ou inventions. Ainsi, tant les approximations "de Vylder" et "diffusion" que les nouvelles approximations "gamma" et "gamma process" sont expliquées et analysées. Ces nouvelles approximations semblent donner de meilleurs résultats dans certains cas.

Traffic Management Strategies for Merge Area in Rural Interstate Work Zones, CTRE 97-12, 1999

The Iowa Department of Transportation, like many other state transportation agencies, is experiencing growing congestion and traffic delays in work zones on rural interstate highways. The congestion results in unproductive and wasteful delays for both motorists and commercial vehicles. It also results in hazardous conditions where vehicle stopped in queues on rural interstate highways are being approached by vehicles upstream at very high speeds. The delays also result in driver frustration, making some drivers willing to take unsafe risks in an effort to bypass delays. To reduce the safety hazards and unproductive delays of congested rural interstate work zones, the Iowa Department of Transportation would like to improve its traffic management strategies at these locations. Applying better management practices requires knowledge of the traffic flow properties and driver behavior in and around work zones, and knowledge of possible management strategies. The project reported here and in a companion report documents research which seeks to better understand traffic flow behavior at rural interstate highway work zones and to estimate the traffic carrying capacity of work zone lane closures. In addition, this document also reports on technology available to better manage traffic in and around work zones.

Stratégies thérapeutiques pour le traitement des gliomes [Therapeutic strategies for the management of gliomas]

The management of gliomas remains challenging and requires a multidisciplinary approach that involves neurosurgeons, radiation therapists and oncologists. For patients with glioblastomas, progress has been made in recent years with the introduction of a combined modality treatment associating radiation therapy and concomitant chemotherapy with the novel alkylating agent temozolomide. This combination resulted in a significant prolongation of survival and increase in the number of patients with survival well beyond two years. Since then, interest in developing new agents in this disease has dramatically increased. In parallel, molecular markers, such as methylation status of MGMT or identification of the translocation of 1p and 19q in oligodendrogliomas have allowed to identify distinct subtypes with exquisite response to treatment or different prognosis. These developments have implications for the development of clinical trials of new potential drug treatments. In this article, we provide a review of the current management of low- and high-grade gliomas, including astrocytomas, oligodendrogliomas and glioblastomas and provide an outlook into future potential therapies.

Optimization strategies for fast detection of positive selection on phylogenetic trees.

MOTIVATION: The detection of positive selection is widely used to study gene and genome evolution, but its application remains limited by the high computational cost of existing implementations. We present a series of computational optimizations for more efficient estimation of the likelihood function on large-scale phylogenetic problems. We illustrate our approach using the branch-site model of codon evolution. RESULTS: We introduce novel optimization techniques that substantially outperform both CodeML from the PAML package and our previously optimized sequential version SlimCodeML. These techniques can also be applied to other likelihood-based phylogeny software. Our implementation scales well for large numbers of codons and/or species. It can therefore analyse substantially larger datasets than CodeML. We evaluated FastCodeML on different platforms and measured average sequential speedups of FastCodeML (single-threaded) versus CodeML of up to 5.8, average speedups of FastCodeML (multi-threaded) versus CodeML on a single node (shared memory) of up to 36.9 for 12 CPU cores, and average speedups of the distributed FastCodeML versus CodeML of up to 170.9 on eight nodes (96 CPU cores in total).Availability and implementation: ftp://ftp.vital-it.ch/tools/FastCodeML/. CONTACT: selectome@unil.ch or nicolas.salamin@unil.ch.

Public health impact of statin prescribing strategies based on JUPITER Modeling study from a population-based study.

OBJECTIVE: To evaluate the public health impact of statin prescribing strategies based on the Justification for the Use of Statins in Primary Prevention: An Intervention Trial Evaluating Rosuvastatin Study (JUPITER). METHODS: We studied 2268 adults aged 35-75 without cardiovascular disease in a population-based study in Switzerland in 2003-2006. We assessed the eligibility for statins according to the Adult Treatment Panel III (ATPIII) guidelines, and by adding "strict" (hs-CRP≥2.0mg/L and LDL-cholesterol <3.4mmol/L), and "extended" (hs-CRP≥2.0mg/L alone) JUPITER-like criteria. We estimated the proportion of CHD deaths potentially prevented over 10years in the Swiss population. RESULTS: Fifteen % were already taking statins, 42% were eligible by ATPIII guidelines, 53% by adding "strict", and 62% by adding "extended" criteria, with a total of 19% newly eligible. The number needed to treat with statins to avoid one CHD death over 10years was 38 for ATPIII, 84 for "strict" and 92 for "extended" JUPITER-like criteria. ATPIII would prevent 17% of CHD deaths, compared with 20% for ATPIII+"strict" and 23% for ATPIII + "extended" criteria (+6%). CONCLUSION: Implementing JUPITER-like strategies would make statin prescribing for primary prevention more common and less efficient than it is with current guidelines.

Stratégies destinées a optimiser l'utilisation des antibiotiques en réanimation [Strategies targeted at optimising antimicrobial therapy in critically ill patients]

Antibiotics are widely used in critical care and adequate empirical treatments has a significant impact on the outcome of many patients. Most nosocomial infections may be due to multidrug-resistant pathogens and requires empirical borad spectrum coverage before identification of the etiologic agents. This is associated with overuse of antibiotics which contributes to the further increase in multidrug-resistances. In this context, new strategies targeted at antibiotic control, such as guidelines and de-escalation are needed to control this evolution.

Pasture characteristics of Italian ryegrass and milk production under different management strategies

The objective of this work was to assess the effects of the sward structure of Italian ryegrass (Lolium multiflorum), during the first grazing cycle, on its morphological and bromatological characteristics throughout the growing season, and on the performance of dairy cows. The treatments consisted of two structures obtained as a function of canopy-light interception: high-light interception (HLI) and low-light interception (LLI), with different pre-grazing heights in the first grazing cycle. Pasture was managed under rotational grazing with a herbage allowance not below 30 kg dry matter (DM) per cow per day. Three grazing cycles, with a grazing interval of 30 days, were evaluated. Pre-grazing herbage mass was greater (2,240 vs. 1,656 kg ha-1 DM), but the proportion of leaf blades was smaller (0.35 vs. 0.43) for HLI swards. Neutral detergent fiber (NDF) content and organic matter digestibility (OMD) were similar between treatments in the first grazing cycle, but in the second and third ones NDF was greater, and OMD lower, for the HLI swards. Milk yields were greater for cows grazing LLI swards (19.4 vs. 21.1 kg per day). Initial grazing with 90% of light interception promotes greater nutritional value in the subsequent cycles.

Gambling against neglect: unconscious spatial biases induced by reward reinforcement in healthy people and brain-damaged patients.

Orienting attention in space recruits fronto-parietal networks whose damage results in unilateral spatial neglect. However, attention orienting may also be governed by emotional and motivational factors; but it remains unknown whether these factors act through a modulation of the fronto-parietal attentional systems or distinct neural pathways. Here we asked whether attentional orienting is affected by learning about the reward value of targets in a visual search task, in a spatially specific manner, and whether these effects are preserved in right-brain damaged patients with left spatial neglect. We found that associating rewards with left-sided (but not right-sided) targets during search led to progressive exploration biases towards left space, in both healthy people and neglect patients. Such spatially specific biases occurred even without any conscious awareness of the asymmetric reward contingencies. These results show that reward-induced modulations of space representation are preserved despite a dysfunction of fronto-parietal networks associated with neglect, and therefore suggest that they may arise through spared subcortical networks directly acting on sensory processing and/or oculomotor circuits. These effects could be usefully exploited for potentiating rehabilitation strategies in neglect patients.

Implicit methods for qualitative modeling of gene regulatory networks.

Advancements in high-throughput technologies to measure increasingly complex biological phenomena at the genomic level are rapidly changing the face of biological research from the single-gene single-protein experimental approach to studying the behavior of a gene in the context of the entire genome (and proteome). This shift in research methodologies has resulted in a new field of network biology that deals with modeling cellular behavior in terms of network structures such as signaling pathways and gene regulatory networks. In these networks, different biological entities such as genes, proteins, and metabolites interact with each other, giving rise to a dynamical system. Even though there exists a mature field of dynamical systems theory to model such network structures, some technical challenges are unique to biology such as the inability to measure precise kinetic information on gene-gene or gene-protein interactions and the need to model increasingly large networks comprising thousands of nodes. These challenges have renewed interest in developing new computational techniques for modeling complex biological systems. This chapter presents a modeling framework based on Boolean algebra and finite-state machines that are reminiscent of the approach used for digital circuit synthesis and simulation in the field of very-large-scale integration (VLSI). The proposed formalism enables a common mathematical framework to develop computational techniques for modeling different aspects of the regulatory networks such as steady-state behavior, stochasticity, and gene perturbation experiments.

Studies toward new anti-cancer strategies based on alpha-mannosidase inhibition.

Changes in the glycosylation pattern of cellular glycoproteins constitute a hallmark in human cancer and influence tumor progression, suggesting that inhibitors of selected glycosidases may control cancer progression. Following the studies on swainsonine, a natural inhibitor of Golgi alpha-mannosidase II, which highlighted the inhibition of cellular mannosidases as a potential innovative approach for the treatment of cancer, several dihydroxylated pyrrolidines and analogues were developed as new potent inhibitors of alpha-mannosidases II able to induce antiproliferative effects in human cancer cells.

Feeding strategies and energy to protein ratio on tambaqui performance and physiology

The objective of this work was to evaluate the effect of feed deprivation and refeeding with diets containing different energy to protein ratios (E/P) on the performance and physiology of juvenile tambaqui (Colossoma macropomum). A 4x2 factorial arrangement with three replicates was used, with four E/P ratios (11.5, 10.5, 9.5, and 8.5 kcal g-1 digestible energy per protein) and two feeding regimens (with and without deprivation), during 60 days. Fish from the food-deprived group were fasted for 14 days and refed from the fifteenth to the sixtieth day, whereas the remaining fish were fed for 60 days. At the end of the experimental period, weight of fish subjected to food deprivation was lower than that of those continuously fed; however, this condition did not influence the physiological parameters analyzed. Tambaqui fed 11.5 kcal g-1 achieved lower final weight than those fed with the other diets, in both regimens. Among the physiological parameters, only plasma protein presented significant increase in fish fed 8.5 kcal g-1, in both feeding regimens, probably due to the higher dietary protein concentration. These results indicate that fish show a partial compensatory growth, and that 10.5 kcal g-1 can be recommended for the diet of juvenile tambaqui.

Stocking densities and feeding strategies in shrimp and tilapia polyculture in tanks

The objective of this work was to evaluate the performance of Pacific marine shrimp (Litopenaeus vannamei) and tilapia (Oreochromis niloticus), in a polyculture in tanks subjected to different stocking densities and feeding strategies, in comparison with monoculture. Two experiments were performed, at the same time, in a completely randomized design with three treatments and four replicates each. Treatments for experiment I were: monoculture with 10 shrimp per m² (10S:0T); polyculture with 10 shrimp and 0.5 tilapia per m² (10S:0.5T); and polyculture with 10 shrimp and 1 tilapia per m² (10S:1T). Shrimp was the main crop, and feed was provided based on shrimp biomass. Treatments for experiment II were: monoculture with 2 tilapia per m² (2T:0S); polyculture with 2 tilapia and 2.5 shrimp per m² (2T:2.5S); and polyculture with 2 tilapia and 5 shrimp per m² (2T:5S). Tilapia was the main crop, and feed was provided based on fish requirements. In the experiment I, tilapia introduction to shrimp culture resulted in lower shrimp growth and poor feed conversion rate. In experiment II, shrimp introduction to tilapia culture did not interfere with fish performance. Polyculture is more efficient with the combination of 2 tilapia and 2.5 or 5 shrimp per m² and feed based on fish requirements.