In this thesis we focus on modeling stochastic mortality and parameter risk in assessing mortality. We explore two mortality stochastic models for modeling the number of deaths in portfolio which consist of one or more than one cohort. We define the term mixture of distributions and introduce Beta-Binomial and Poisson-Gamma model. We address immediate life annuities and we apply Bayesian Poisson- Gamma model to quantify longevity risk on data. The obvious increasing trend of average lifetime leads insurance companies to greater protection against longevity risk. We show how to deal with solvency rules by internal models designed consistently with the requirement in the standard formula of Solvency II. Powered by TCPDF (www.tcpdf.org)

Dans cette thèse, nous avons examiné l'impact de la communication des informations de risque dans les rapports annuels sur la liquidité, la réputation, et les risques spécifiques, systématiques et totaux. Nous avons développé une nouvelle méthodologie pour mesurer la communication sur les risques en utilisant six listes de mots (les mots incertains, d’opportunité, négatifs, de forme faible, des règlements juridiques et gouvernementaux, de l'environnement et de responsabilité sociale). En outre, nous avons utilisé plusieurs modèles empiriques (effets fixes, effet aléatoire, MCO, logistique groupée, effets fixes conditionnels logistiques, effets aléatoires logistiques, et le modèle linéaire mixte). L'analyse révèle plusieurs résultats importants. Premièrement, nous avons constaté que la communication des risques conduit à une baisse de liquidité, à une meilleure réputation, à une baisse des risques spécifiques, et à une augmentation des risques systématiques. Les résultats montrent également que la communication des risques n’influence pas la réputation des entreprises à haut risque. D’ailleurs, au cours de la dernière crise financière, la communication des risques a augmenté les risques spécifiques et totaux. Finalement, la communication des risques augmente les risques totaux et les risques systématiques pour les entreprises à haut risque, tandis qu'elle diminue les risques pour les entreprises à faible risque. / This Ph.D. dissertation explores the effect of the communication of risk on several factors in the French market. To measure communication about risk, we used content analysis. We developed a new method of measurement by using several word lists to capture different types of ambiguity and risk reporting (67 environmental and social responsibility words, 889 legal and government-regulation words, 2184 negative words, 306 uncertain words, 25 opportunity words, and 32 weak words). This thesis is organized into three chapters. The first chapter is devoted to studying the effect of risk communication on firm liquidity. The results show that an intense tone of risk and uncertain information in annual reports negatively affect liquidity. In the second chapter, we examine the effect of risk communication on companies’ reputations. We detect that risk reporting positively affects reputation. This result is robust for alternative empirical models (pooled OLS, fixed effects, and random effects) as well as for alternative measurements of reputation. In addition, we explore the risk-reporting behaviors of very high- and low-risk companies. We find that risk-disclosure behavior is sensitive to a company’s level of risk. The third chapter is dedicated to analyzing the effect of risk communication on company risks (unsystematic, systematic, and total risk). We find that risk communication is associated positively with systematic risk, and negatively with unsystematic risk. In contrast, during the financial crisis of 2008, we find a negative association between risk communication on the one hand and unsystematic and total risk on the other hand. Moreover, we observe that high-risk firms will not reduce their risks through more communication about risk.

This paper studies factors that influence the stock return on the Shanghai Stock Exchange (SSE) market. To achieve this goal, a stock-fixed effects model is estimated using a panel data sample comprising 100 companies listed on the SSE market during the 72-month period from January 2002 to December 2007. I find that number of trades and book-to-market value in both up and down markets have a significant and positive impact on stock returns during the studied period, whereas stock returns were negatively affected by systematic risk in both up and down markets although less so in up markets. Price to earnings ratio did not show any significant effect on stock returns on the SSE. My overall results indicate that SSE did not satisfy the efficient market hypothesis 1 during the studied period from January 2002 to December 2007.

The initiative by the EU to combat global warming through the introduction of a cap-and-trade system for greenhouse gas emissions in 2005, known as the EU Emissions Trading System (ETS), resulted in the inception of a new financial market. The right to emit one tonne of CO2-equivalents, as well as derivatives on this right, have become commodities, traded both through exchanges and over the counter. A relevant question thus becomes how a market maker trading these derivatives should hedge their exposures. This thesis examines how stochastic optimisation can be used to hedge a portfolio of futures, forwards, and emissions rights in the EU ETS, while taking into account market microstructre effects such as transaction costs. This is done through the implementation of a Stochastic Programming (SP) model that weights portfolio risk against transaction costs,where the entire term structure of futures is Monte Carlo-simulated. The term structure of the futures is analysed by decomposing futures prices into the spot price, the term structure of the risk free interest rate, and the term structure of the convenience yield, as was first done by Working (1948). These are estimated using the non-parametric optimisation framework of Blomvall (2017), where EUROIS contracts and ICE EUA futures are used as benchmark instruments. It was found that the method results in smooth yield curves with small repricing errors, thanks to suitable parameter calibration through 3-fold Cross Validation for both curves. From day-to-day changes in the resulting curves, three systematic risk factors for each curve, that capture more than 98% of the variance during the analysed period from October 2012 to March 2018, were found with PCA. These factors were then fitted to univariate GARCH-models, and normal mixture copulas. This model allows for the hedging problem to be solved with SP. For the out-of-sample period from March 2018 to April 2022, the results show promise, as the portfolios hedged with SP are considerably less volatile than both a statically hedged and an unhedged portfolio. Furthermore, for some values of the parameter weighting risk and costs, these portfolios yield mean variance efficiency.

To maintain solvency intimes of severe economic downturns banks and financialinstitutions keep capital cushions that reflect the risks in the balance sheet.Broadly,how much capital that is being held is a combination of external requirementsfromregulators and internal assessments of credit risk. We discuss alternatives totheBasel Pillar II capital add-on based on multi-factor models for held capitaland howthese can be applied so that only concentration (or sector) risk affects theoutcome,even in a portfolio with prominent idiosyncratic risk. Further, the stabilityandreliability of these models are evaluated. We found that this idiosyncraticrisk canefficiently be removed both on a sector and a portfolio level and that themulti-factormodels tested converge.We introduce two new indices based on Risk Weighted Assets (RI) and EconomicCapital (EI). Both show the desired effect of an intuitive dependence on the PDand LGD. Moreover, EI shows a dependence on the inter-sector correlation. Inthesample portfolio, we show that the high concentration in one sector could be(better)justified by these methods when the low average LGD and PD of this sector weretaken into consideration. / För att behålla solvens itider av svår lågkonjunktur håller banker och finansiellainstitutioner buffertar med kapital som reflekterar risken i balansräkningen. Istoradrag så är mängden kapital som hålls beroende av en kombination av externa kravfrån regulatorer och interna uppskattningar av kredit risken. Den häravhandlingendiskuterar alternativ till Basel pelare II kapital påslaget som är baserade påmulti-faktor modeller för kapital och hur dessa kan appliceras så att endastkoncentration(eller sektor) risk påverkar resultat, även i en portfölj med tydligidiosynkratiskrisk. Utöver detta behandlas stabilitet och reliabilitet hos dessa modeller.Genomdetta hittas att den idiosynkratisk risk kan effektivt tas bort på bådeportfölj- ochsektornivå och att de multifaktor modeller som testas konvergerar.Den här avhandlingen introducerar två nya index, baserat på Risk WeightedAssets(RI) och Economic Capital (EI). Båda visar på den önskade effekten av ettintuitivtberoende av PD och LGD. Dessutom visar EI ett beroende av inter-sektor korrela-tion. Med stickprovsportföljen som används var det tydligt att högkoncentrationi en sektor kunde (bättre) rättfärdigas av båda dessa metoder då LGD och PD försektorn i fråga har beaktats.

No description available.

本研究透過聯合系統與獨特風險綜合評估違約的強度，假設市場上經濟變數或資訊影響系統之違約強度，然若直接考慮所有經濟變數到模型中將可能會有共線性或維度過高之疑慮，因此透過狀態空間模型來設定狀態變數以及經濟變數之關係並將萃取三大狀態變數分別用以描述市場實質活動面、通貨膨脹以及信用環境。另外，將透過結構式模型來計算獨特性風險大小，當個別潛在的變數低於一定數值將導致個別的違約事件發生。而因布朗運動可能無法描述或校準市場上違約之鋒態以及偏態，將進一步考慮Variance Gamma過程用以更準確描述真實違約狀況。最後透過結合以上兩個風險綜合評估下，考慮一個聯合違約模型來評價信用違約交換之信用價差。 / Systematic and idiosyncratic risks are supposed to jointly trigger the default events. This paper identifies three fundamental risks to capture the systematic movement: real activity, inflation, and credit environment. Since most macroeconomic variables fluctuate together, the state-space model is imposed to extract the three variables from macroeconomic data series. In the idiosyncratic part, the structural model is applied. That is, idiosyncratic default is triggered by the crossing of a barrier. For improvement of the underlying lognormal distribution, we assume the process for the potential variable of the firm follows a Variance Gamma process, sufficient dimensions of which can fit the skewed and leptokurtic distributions. Under the specific setting of combinations of the two risks (the so-called joint default model), we price credit default swaps.

Cette thèse porte sur la prédiction de la détresse financière et son impact sur le rendement des actions. L’objet principal de cette thèse est de : (i) prédire la détresse financière des petites et moyennes entreprises françaises en utilisant plusieurs spécifications économétriques tels que, le modèle Logit, les réseaux de neurones artificiels, la méthode SVM et la régression des moindres carrés partiels, et (ii) d’identifier les facteurs de risque de détresse financière à caractère systématique, explicatifs des rendements des actions, et additionnels au modèle de Fama et French (1993) tels que le momentum, la détresse relative, la liquidité et la Value-at-Risk, sur le marché boursier Français. Cette étude comporte deux parties. La première partie, composée de 2 chapitres, s’interroge sur les principaux indicateurs discriminants entre les petites et moyennes entreprises françaises saines et celles en détresse financière un an et deux ans avant la défaillance. Elle mobilise différentes approches de prédiction et aboutit à des résultats empiriques qui font l’objet d’analyse. La deuxième partie, composée aussi de 2 chapitres, étudie le pouvoir explicatif, du modèle de Fama et French (1993) augmenté de certains facteurs de risque, mais aussi des modèles alternatifs à cette approche dans le contexte français. Les tests portent aussi sur le caractère systématique des facteurs de risque additionnels ou alternatifs, explicatifs des rendements des actions. Les résultats empiriques obtenus font l’objet d’analyse et permettent de proposer des implications managériales aux décideurs. / This thesis focuses on financial distress and its impact on stock returns. The main goal of this dissertation is: (i) to predict the financial distress of French small and medium-sized firms using a number of techniques namely Logit model, Artificial Neural Networks, Support Vector Machine techniques, and Partial Least Squares, and (ii) to identify the systematic risk factors of financial distress that can explain stock returns, in addition to those of Fama and French (1993) such as the momentum, the relative distress, the liquidity, and the Value-at-Risk in the French stock market. This study has been concretized in two parts. The first part, composed of 2 chapters, wonders about the main indicators that can discriminate between distressed and non-distressed French small and medium-sized firms one and two years before default. It mobilizes different prediction techniques and leads to the empirical results that are the subject of the analysis. The second part, composed also of 2 chapters, investigates the explanatory power of Fama and French (1993) model augmented by a number of risk factors, as well as alternative models in the French context. The tests also focus on the systematic nature of the additional or alternative risk factors, explaining the stock returns. The obtained empirical results are analyzed and propose managerial implications to decision makers.

Le management des risques de projet est une préoccupation croissante dans le domaine de la construction. Il ne se limite pas seulement à l’analyse des risques techniques, mais couvre aussi les risques financiers, économiques, organisationnels, réglementaire, contractuel, et d’autres types de risques cruciaux pour des projets de construction complexes et stratégiques. Le management des risques nécessite l’identification, l’analyse, le suivi des risques et des opportunités pendant tout le cycle de vie du projet. Le processus formalisé et systémique de management des risques pour des projets de construction complexes et stratégiques permet d’identifier et d’analyser les risques attachés d’une part à la décomposition chronologique du projet (phases, sous-phases, tâches), à la décomposition organisationnelle du projet (aux acteurs projet), aux ressources, aux contrats, aux facteurs externes et d’autre part aux relations entre ces éléments. Le processus formalisé et systémique s’adapte au caractère dynamique et évolutif du projet, au type de contrat et au type de projet, au niveau de détail souhaité et à la vision de l’acteur qui fait le management des risques. L’ensemble de la démarche est fortement nourri de l’expérience tirée de projets réels au sein de l’entreprise partenaire. Un outil de gestion des risques est mis en place pour mettre en pratique la démarche théorique et pour tester la méthodologie proposée sur plusieurs études de cas des projets Partenariat Public Privé (PPP) et Conception-Construction-Maintenance. / Project risk management is a growing concern in the field of construction. It is not limited to technical risks, but also covers financial risks, economic, organizational or contractual and any type of risks crucial for complex and strategic construction projects. Risk management process involves the identification, analysis, monitoring of risks and opportunities throughout the project life cycle. The formalized and systematic risk management process first identifies and analyzes the risks associated with the chronological decomposition of the project (phases, sub-phases, tasks), with the organizational structure of the project (project actors), with resources, contracts, external factors and material or immaterial flows between these elements. The formalized and systematic approach adapts to the dynamic and evolving nature of the project, to the type of contract and the type of project, to the level of detail and the vision of the stakeholder who manages risks. The whole process is highly fed by real projects study cases. A tool for risk management is developed to put into practice the theoretical approach and to test the process in the case studies of Public Private Partnership (PPP) and Design-Build-Maintenance projects.

本文主要研究動態回復率之建構。並搭配使用機率勺斗法，將資產之離散損失分配建構出合成型擔保債權憑證分劵損失分配。歸納出離散動態回復率對合成型擔保憑證分劵之風險承擔與信用價差變化。本文發現在動態回復率中，即使在相同條件下有一樣預期損失，能使其債權群組損失分配之標準差較固定回復率小，且可使投資組合巨額損失部份產生厚尾分配現象。動態回復率對各分劵面臨共同存活與違約機率具有緩和或增強分劵承擔風險之作用。在單因子高斯連繫結構靜態違約下，透過隨機回復率能增加動態系統性風險因子之描繪。類似於將系統風險因子分配由標準常態分配改成t分配或是債權群組間違約相關係提高。