Background: Stress tests have been increasingly used as a part of the supervisory tool by national regulators after the financial crisis, which can also be used to conduct authorities’ supervisory for determining bank capital levels, assessing the health of a bank. Purpose: The main purpose of this study is to assess whether some micro factors play important roles on the changes of Common Equity Tier One Capital Ratio (between the bank accounting value and the stress testing results under the adverse scenarios).  Our secondary purpose is to investigate if our empirical results will help to provide some theoretical suggestions to regulators when they exercise stress tests.   Method: An empirical analysis by using Panel Data, introducing GARCH model to measure volatility.   Empirical foundation: The results of EU-wide stress tests and bank financial statements   Conclusion: The coefficient associated with non-performing loans to total loans is positively significant and the coefficient associated with bank size is negatively significant.  In addition, the financial system of strong banks is better to absorb financial shocks. These findings are useful, as banks is a reflection of the financial stability of an economic entity, we can use these findings as another reason to pay attention to the process of the stress testing rather just stress testing results.

L’objet de cette thèse est d’analyser d’un point de vue financier la cotation et le comportement des obligations privées à la Bourse de Paris à partir de 1838 jusqu’à l’éclatement de la Première Guerre mondiale. Cette étude est divisée en deux parties : La première relate la création et l’évolution des émissions obligataires (marché primaire) durant le 19ème siècle. On s’intéresse aux grands émetteurs qui ont su se servir le mieux des obligations et les raisons de leur succès. Dans cette partie seront également traitées les caractéristiques techniques et les innovations financières des émissions. La deuxième partie tente d’analyser le comportement boursier des obligations(marché secondaire).Pour cela un indice général des cours d’obligations durant le 19ème siècle a été calculé. En se basant sur cet indice nous mettons en lumière pour la première fois, les caractéristiques de ce marché (rentabilité, volatilité, …). Cela permet de comparer nos résultats avec ceux des études antérieures sur les marchés d’actions et de la rente au 19ème siècle. Cet indice permet également de tester les diverses hypothèses financières relevant de la théorie financière moderne (efficience informationnelle, cointégration avec des autres compartiments du marché,…). / This thesis studies the French corporate bonds market during the 19th century. Despite its importance the performance of the corporate bonds quoted on the Paris Bourse has never been studied. In order to analyse this market, a price index of the corporate bond market has been created by using modern techniques. The creation of the index was made possible thanks to an original database created by new data, which has never been used before and collected directly from the publications of the market authorities during the nineteenth century. Thanks to the index, the risk and the return of the market have been measured. Then we compared the performance of the French corporate bonds with those of the stocks and government bonds; the results of thecomparisons are interesting. This study demonstrates that the corporate bonds are the least risky securities and their rate of return is higher than the government bonds during the nineteenth century. Some econometric tests have also been used to compare the efficiency of bond market with the other segments of the Paris Bourse.

Este trabalho analisa e modela a Estrutura a Termo das Taxas de Juros objetivando ao teste da Hipótese das Expectativas(HE) na ponta curta da curva de juros e a uma aplicação da teoria de Markowitz (1952) no mercado de renda fixa utilizando a estrutura proposta por Caldeira, Moura e Santos (2015). Para estes fins foram utilizados dados dos contratos futuros de 1 dia dos depósito interbancários (DI1) negociados na BMF interpolados em maturidades fixas, sendo utilizados em base semanal quando do teste da HE e em base diária para a construção dos portfólios de mínima variância. Os resultados encontrados para o teste da HE sugerem a invalidade da teoria, uma vez que o prêmio de risco é se mostra ajustável a um modelo GARCH-M e, portanto, variante no tempo. Os portfólios de mínima variância ajustados nas versões irrestrita e restrita (duration máxima de 1 ano) se mostraram consistentes, tendo superado quase a totalidade dos fundos analisados. O portfólio de mínima variância irrestrito obteve o maior Índice de Sharpe no período analisado. / This work analyzes and model the Term Structure of Interest Rates seeking testing Expectation Hypothesis in the short end of the Yield Curve and to apply the portfolio theory to the fixed income context using the framework proposed by Caldeira, Moura e Santos (2015). We used a database of constant maturities interbank deposits’s future contracts. The results suggest Expectation Hypothesis doesn’t hold and risk premium could be modeled by a GARCH-M framework, being time variant. The bond portfolio optimized were, in general, consistent with high sharpe ratio relative to other funds and beated the chosen benchmark during the period analyzed.

Cette thèse a pour vocation principale d’expliquer l’énigme de la prime de risque en évaluant l’impact de la crise financière récente des Subprimes aux Etats-Unis sur la prime de risque. Etant donné que le coefficient du risque systématique demeure le facteur majeur de la prime de risque, nous avons recouru à une version conditionnelle du MEDAF afin d’étudier l’impact de la crise sur le bêta conditionnelle en utilisant comme spécification économétrique le GARCH multivarié (BEKK) d’Engel et Kroner (1995). La crise des subprimes qui a débuté à la fin de l’année 2007 avec l’effondrement du marché immobilier américain, a d'abord eu des conséquences sur la sphère financière locale. Mais elle a pris de l'ampleur, diffusant aux autres grandes places financières. Depuis longtemps, l’aversion au risque n’avait pas été aussi élevée et la faillite de la banque d’affaires Lehman Brothers en mi-septembre, a constitué le point d’orgue de la crise. Dans ce travail, nous nous sommes basé sur le marché boursier français et nous avons essayé de détecter l’impact de la crise financière sur la prime de risque des indices sectoriels. Pour ce faire, nous avons essayé de montrer l’effet de la crise sur le risque systématique bêta des indices. D’abord, nous avons créé des bêtas journaliers par le modèle GARCH bivarié (BEKK) d’Engel et Kroner sur la période allant de 01 Janvier 2007 au 31 Décembre 2014. Ensuite, nous avons régressé le bêta en introduisant comme variables explicatives la variance conditionnelle de l’indice et la variance conditionnelle du marché. Des variables Dummy y sont introduites pour détecter l’effet de la crise sur le bêta à partir du 15 Septembre 2008.Nos résultats ont montré que la crise financière a provoqué une hausse de volatilité des indices sectoriels. Cette hausse a eu un impact direct sur le risque systématique bêta ( positif significatif). Celle-ci a augmenté pour la majorité des secteurs durant la période après crise. / This thesis aims primarily to explain the enigma of the risk premium by assessing the impact of the recent financial subprime crisis in the United States on the risk premium. Since the systematic risk factor remains the major factor in the risk premium, we used a conditional version of the CAPM to study the impact of the crisis on the conditional beta using econometric specification as the multivariate GARCH ( BEKK) Engel and Kroner (1995). The subprime crisis that started in late 2007 with the collapse of the US housing market, first had an impact on the local financial sector. But it gained momentum, spreading to other major financial centers. Long time, risk aversion was not as high and the bankruptcy of investment bank Lehman Brothers in mid-September, was the organ of the crisis developed. In this work, we based on the French stock market and we tried to detect the impact of the financial crisis on the risk premium sector indices. To do this, we tried to show the effect of the crisis on the systematic risk beta indices. First, we created daily by the beta bivariate GARCH (BEKK) Engel and Kroner on the period from January 1, 2007 to December 31, 2014. Then we dropped the beta introducing explanatory variables variance conditional index and the conditional variance of the

The thesis compares GARCH volatility models and Stochastic Volatility (SV) models with Student's t distributed errors and its empirical forecasting per- formance of Value at Risk on five stock price indices: S&P, NASDAQ Com- posite, CAC, DAX and FTSE. It introduces in details the problem of SV models Maximum Likelihood examinations and suggests the newly devel- oped approach of Efficient Importance Sampling (EIS). EIS is a procedure that provides an accurate Monte Carlo evaluation of likelihood function which depends upon high-dimensional numerical integrals. Comparison analysis is divided into in-sample and out-of-sample forecast- ing performance and evaluated using standard statistical probability back- testig methods as conditional and unconditional coverage. Based on empirical analysis thesis shows that SV models can perform at least as good as GARCH models if not superior in forecasting volatility and parametric VaR. 1

O conhecimento do risco de ativos financeiros é de fundamental importância para gestão ativa de carteiras, determinação de preços de opções e análise de sensibilidade de retornos. O risco é medido através da variância estatística e há na literatura diversos modelos econométricos que servem a esta finalidade. Esta pesquisa contempla o estudo de modelos determinísticos de volatilidade, mais especificamente os modelos GARCH simétricos e assimétricos. O período de análise foi dividido em dois: de janeiro de 2000 à fevereiro de 2008 e à outubro de 2008. Tal procedimento foi adotado procurando identificar a influência da crise econômica originada nos EUA nos modelos de volatilidade. O setor escolhido para o estudo foi o mercado de petróleo e foram escolhidas as nove maiores empresas do setor de acordo com a capacidade produtiva e reservas de petróleo. Além destas, foram modeladas também as commodities negociadas na Bolsa de Valores de Nova York: o barril de petróleo do tipo Brent e WTI. A escolha deste setor deve-se a sua grande importância econômica e estratégica para todas as nações. Os resultados encontrados mostraram que não houve um padrão de modelo de volatilidade para todos os ativos estudados e para a grande maioria dos ativos, há presença de assimetria nos retornos, sendo o modelo GJR (1,1) o que mais prevaleceu, segundo a modelagem pelo método da máxima verossimilhança. Houve aderência, em 81% dos casos, dos ativos a um determinado modelo de volatilidade, alterando apenas, como eram esperados, os coeficientes de reatividade e persistência. Com relação a estes, percebe-se que a crise aumentou os coeficientes de reatividade para alguns ativos. Ao se compararem as volatilidades estimadas de curto prazo, percebe-se que o agravamento da crise introduziu uma elevação média de 265,4% em relação ao período anterior, indicando um aumento substancial de risco. Para a volatilidade de longo prazo, o aumento médio foi de 7,9%, sugerindo que os choques reativos introduzidos com a crise, tendem a ser dissipados ao longo do tempo. / The knowledge of the risk of financial assets is of basic importance for active management of portfolios, determination of prices of options and analysis of sensitivity of returns. The risk is measured through the variance statistics and has in literature several econometrical models that serve to this purpose. This research contemplates the study of deterministic models of volatility, more specifically symmetrical and asymmetrical models GARCH. The period of analysis was divided in two: January of 2000 to the February of 2008 and the October of 2008. Such a proceeding was adopted trying to identify the influence of the economic crisis given rise in U.S.A. in the volatility models. The sector chosen for the study was the oil market and had been chosen the nine bigger companies of the sector in accordance with the productive capacity and reserves of oil. Beyond these, there were modeled also the commodities negotiated in the Stock Exchange of New York: the barrel of oil of the types Brent and WTI. The choice of this sector is due to his great economical and strategic importance for all the nations. The results showed that there was no a standard of model of volatility for all the studied assets and for the majority of them, there is presence of asymmetry in the returns, being the model GJR (1,1) that more prevailed, according to the method of likelihood. There was adherence, in 81 % of the cases, of the assets to a determined model of volatility, altering only the coefficients of reactivity and persistence. Regarding these, it is realized that the crisis increased the coefficients of reactivity for some assets. In relation to the volatilities of short term, it is realized that the aggravation of the crisis introduced an elevation of 265,4% regarding the previous period, indicating a substantial increase of risk. In relation to the volatility of long term, the increase was 7,9 %, suggesting that the reactive shocks introduced with the crisis have a tendency to be dispersed along the time.

Neste trabalho, aprofundamos o estudo sobre risco sistêmico via funções de agregação. Consideramos três carteiras diferentes como proxy para um sistema econômico, estas carteiras são consistidas por duas funções de agregação, baseadas em todos as ações do E.U.A, e um índice de mercado. As medidas de risco aplicadas são Value at Risk (VaR), Expected Shortfall (ES) and Expectile Value at Risk (EVaR), elas são previstas através do modelo GARCH clássico unido com nove funções de distribuição de probabilidade diferentes e mais por um método não paramétrico. As previsões são avaliadas por funções de perda e backtests de violação. Os resultados indicam que nossa abordagem pode gerar uma função de agregação adequada para processar o risco de um sistema previamente selecionado. / In this work, we deepen the study of systemic risk measurement via aggregation functions. We consider three different portfolios as a proxy for an economic system, these portfolios are consisted in two aggregation functions, based on all U.S. stocks and a market index. The risk measures applied are Value at Risk (VaR), Expected Shortfall (ES) and Expectile Value at Risk (EVaR), they are forecasted via the classical GARCH model along with nine distribution probability functions and also by a nonparametric approach. The forecasts are evaluated by loss functions and violation backtests. Results indicate that our approach can generate an adequate aggregation function to process the risk of a system previously selected.

This thesis has treated the subject of DCC-GARCH model’s forecasting ability and Value-at- Risk applications on the Scandinavian foreign exchange market. The estimated models were based on daily opening foreign exchange spot rates in the period of 2004-2013, which captured the information in the financial crisis of 2008 and Eurozone crisis in the early 2010s. The forecasts were performed on a one-day rolling window in 2014. The results show that the DCC-GARCH model accurately predicted the fluctuation in the conditional correlation, although not with the correct magnitude. Furthermore, the DCC-GARCH model shows good Value-at-Risk forecasting performance for different portfolios containing the Scandinavian currencies.

自從Engle（1982）觀察金融資產報酬序列具有波動叢聚的現象，進而提出自迴歸條件變異數異質（Autoregressive Conditional Heteroskedasticity）模型後，解決了傳統的時間序列模型如自我迴歸移動平均（Autoregressive Moving Average；簡稱ARMA）模型在財務金融的實證研究上對變異數異質（Heteroskedasticity）的現象不能做有效解釋的問題後。陸續的延申模型如Bollerslev（1986）一般化自我迴歸條件異質變異（General Autoregressive Conditional Heteroskedasticity；簡稱GARCH）模型、Chou（1988）的GARCH - M（General Autoregressive Conditional Heteroskedasticity in Mean；簡稱GARCH - M）模型，已廣泛的應用於分析股票市場股價持續波動的問題上。 同時，由於國際股市間通常存在著訊息傳遞的現象，國際股市波動命題已在學術界廣泛的被討論，因此本文選取與台灣經濟景氣息息相關的美國、日本來分析訊息的傳遞型式，並且討論風險貼水項在解釋股價指數報酬上是否具有解釋能力。此外，若美、日、台三地的股價指數無法擊敗隨機漫步模型Ⅰ，則三地股票市場是效率市場，股價指數將是完全隨機不可預測的，因此本文經由BDS檢定方法來檢定隨機漫步模型Ⅰ是否足以表徵三地的價格過程，做為在是否進行多變數GARCH模型估計的參考。同時，若市場間的非同步現象嚴重的話將對估計結果產生影響，因此，本文也對三地股市非同步現象加以探討，視其是否有非同步調整的必要再加以調整，以避免估計時產生誤差。 接著，本文亦就股價指數歷史資料橫斷面探討國際間股市的互動性，經由Tank and Kwok（1997）模型加以延申，進行台灣、美國、日本國際投資組合多角化效益之分析，探討的命題主要是週末效果、元月效果，和Rogalski效果是否顯著存在於台灣、美國、日本股票市場，即分析上述效果是否對國際投資組合均報酬、報酬波動、風險分散利益產生影響。 總言之，經由上述的實證過程，將使我們對美、日、台股市具有更進一步的了解。同時，三地股票市場訊息的連結與傳遞型式分析，也可做為投資人在進行股票市場投資與國際投資組合建構時的決策參考。

本文主要探討出售長期資產對公司股價的影響,由於金融性資產時間序列資料有異質變異數現象,傳統事研究以最小平方法無法正確描述殘差項之變異,故使用一般自我迴歸異質變異數(GARCH)模型.由實証發現,市場對公司出售資產的反應視其處分目的而定,對為了改善營運而出售資產的公司,一般相信公司營運能提升,有助於增進公司價值;對為了改善財務而出售資產的公司,由於公司本身財務狀況不佳,雖藉出售資產取得資金,但市場反應仍不佳.另外,由迴歸分析也發現處分目的為宣告期間累積異常報酬重要的解釋變數.