Field of examination is theoretical and empirical review of dynamic CAPM models that assume non constant volatility and correlation. In other words time evolution is considered in estimation process. As theoretical basement is recommended to be R. Engle's (Dynamic Conditional Beta) research and other sources.

Model CAPM je považován za základní model při oceňování systematického risku aktiv a jeho provázanosti s výnosností trhu. Tato práce využívá této struktury a použitím různých metod, mezi které patří OLS, DCC MGARCH a SSF modelovaní, se snaží najít nejvhodnější metodu z výše zmíněných, která dokáže nejlépe odhadnout koeficienty systematického risku. Tyto koeficienty jsou dále použity pro zajištění rizika portfolií, které jsou vytvořeny z akcií obchodovaných na různých burzách- NYSE Composite a NASDAQ Composite. Na základě obdržených výsledků o výkonu zajištění rizika v každém portfoliu budeme schopni vyhodnotit, která z metod je nejvhodnější pro odhad systematické risku v modelu CAPM. Klíčová slova: CAPM, Systematický risk, Portfolio risk hedge, OLS, DCC MGARCH, SSF model JEL Classification: C22, C58, G11, G12, G15 Author's e-mail: danielsopov@email.cz Supervisor's e-mail: andrlikova@gmail.com

Modeling and forecasting the volatilities of high-frequency data observed on the prices of financial assets are vibrant research areas in econometrics and statistics. However, most of the available methods are not directly applicable when the number of assets involved is large, due to the lack of accuracy in estimating high-dimensional matrices. This paper compared two methodologies of vast volatility matrix estimation for high-frequency data. One is to estimate the Average Realized Volatility Matrix and to regularize it by banding and thresholding. In this method, first we select grids as pre-sampling frequencies,construct a realized volatility matrix using previous tick method according to each pre-sampling frequency and then take the average of the constructed realized volatility matrices as the stage one estimator, which we call the ARVM estimator. Then we regularize the ARVM estimator to yield good consistent estimators of the large integrated volatility matrix. We consider two regularizations: thresholding and banding. The other is Dynamic Conditional Correlation(DCC) which can be estimated for two stage, where in the rst stage univariate GARCH models are estimated for each residual series, and in the second stage, the residuals are used to estimate the parameters of the dynamic correlation. Asymptotic theory for the two proposed methodologies shows that the estimator are consistent. In numerical studies, the proposed two methodologies are applied to simulated data set and real high-frequency prices from top 100 S&P 500 stocks according to the trading volume over a period of 3 months, 64 trading days in 2013. From the perfomances of estimators, the conclusion is that TARVM estimator performs better than DCC volatility matrix. And its largest eigenvalues are more stable than those of DCC model so that it is more approriable in eigen-based anaylsis.

Le développement du système nerveux central fait intervenir de nombreux processus cellulaires comme la prolifération, la différenciation, l'apoptose, la migration neuronale et le guidage axonal. Ces processus sont finement régulés et aboutissent à la formation d'un système nerveux central fonctionnel. Au cours de cette thèse, je me suis principalement intéressé à une molécule chimiotropique, la Nétrine-1 et à ses récepteurs APP (Amyloïd-b Precursor Protein) et DCC (Deleted in Colorectal Cancer). La Nétrine-1 est impliquée à la fois dans le contrôle de la navigation neuronale (ie migration neuronale et guidage axonal) et dans celui de la survie neuronale. Un premier aspect de mon travail a consisté à étudier le rôle d'un nouveau récepteur à la Nétrine-1 : APP. Nous avons donc démontré que APP est un récepteur à la Nétrine-1 impliqué dans le guidage des axones commissuraux. En effet, APP collabore avec DCC afin de stimuler la croissance axonale en réponse à la Nétrine-1. Le second aspect de mon travail a été l'étude du rôle de la Nétrine-1 en tant que facteur de survie neuronale. En effet, son récepteur DCC appartient à la famille des récepteurs à dépendance. En absence de Nétrine-1, DCC ne reste pas inactif, mais déclenche une signalisation pro-apoptotique qui va éliminer la cellule. Nous avons démontré que pendant le développement du système nerveux, la Nétrine-1 inhibe cette signalisation et contrôle la survie des neurones commissuraux. Le contrôle de la signalisation pro-apoptotique de DCC par la Nétrine-1 pourrait ainsi réaliser un contrôle qualité des projections axonales et faciliter l'orientation du cône de croissance lors du guidage du cône de croissance / The development of the central nervous system requires several cellular processes such as proliferation, differentiation, apoptosis, neuronal migration and axon guidance. This whole process is finely regulated and leads to the formation of a functional central nervous system. During my thesis, I have mainly worked on a chemotropic molecule, Netrin-1 and its receptors APP (Amyloid-ß Precursor Protein) and DCC (Deleted in Colorectal Cancer). Netrin-1 is involved in both neuronal navigation (neuronal migration and axonal guidance) and neuronal survival. I have firstly investigated the role of a new Netrin-1 receptor, APP during axonal guidance. We have shown that APP is involved in the guidance of commissural axons. Indeed, APP collaborates with DCC in order to promote axonal outgrowth in response to Netrin-1. Thereafter, I have investigated the role of Netrin-1 as a neuronal survival factor. In fact, the Netrin-1 receptor DCC belongs to the dependence receptor family. In absence of Netrin-1, DCC does not remain inactive, since it triggers a pro-apoptotic signalling pathway. During the development of the central nervous system, we have shown that Netrin-1 blocks the pro-apoptotic signalling pathway and promotes neuronal survival. The Netrin-1 control of DCC pro-apoptotic signalling might carry out a “quality control” of the axonal projection or/and it could promote the steering of the growth cone during axonal guidance

DCC est un récepteur transmembranaire ayant pour ligand la nétrine-1. DCC appartient à la famille des récepteurs à dépendance, qui ont la particularité de ne pas rester inactifs en absence de ligand mais, au contraire, d'induire activement l'apoptose lorsqu'ils sont dans un état non lié, ce qui explique leur fonction suppresseur de tumeur. Mon travail de thèse s'est articulé autour de deux axes de recherche: Axe 1 : Implication du couple DCC/nétrine-1 dans la lymphomagenèse. Nous avons montrer que l'expression du couple DCC/nétrine-1 est dérégulée dans les lymphomes du manteau (LM) ainsi que dans les lymphomes diffus à grandes cellules B (LDGC-B) : l'expression de la nétrine-1 est augmentée dans les LM et les LDGC-B de type ABC, alors que l'expression de DCC est diminuée dans les LDGC-B de type GC. Sur le plan thérapeutique, rétablir l'apoptose induite par DCC en bloquant l'interaction avec son ligand nétrine-1 induit une diminution du volume tumoral dans un modèle de xénogreffe. Le couple DCC/nétrine-1 pourrait également jouer un rôle dans le développement des lymphomes extranodaux. Axe 2 : Caractérisation fonctionnelle de la protéine ADAMTSL-1.L'orthologue de DCC chez C. elegans, UNC-40 interagit fonctionnellement avec la protéine MADD-4 lors de la mise en place des jonctions neuromusculaires. Nous n'avons pas établi l'existence d'une telle connexion entre DCC et l'orthologue de MADD-4 chez les Mammifères, ADAMTSL1. La fonction d'ADAMTSL1 étant inconnue, nous avons utilisé différents systèmes in vitro et in vivo, dont un modèle murin invalidé constitutivement pour ce gène, pour déterminer la fonction de cette protéine matricielle, notamment dans les tissus musculaires et cartilagineux / DCC is a tramsmembrane protein, receptor of netrin-1. DCC belongs to the dependence receptor family, which have a dual functionality. Indeed, they are not inactive when not bound by their ligand but instead they actively induce apoptosis thus explaining that they are tumor suppressors. My PhD project contains 2 axes :Axe 1 : Involvement of DCC/netrin-1 during lymphomagenesis. We demonstrated that the gene expression of DCC and netrin-1 is deregulated in mantle cell lymphoma (MCL) and in diffuse large b cell lymphoma (DLBCL) : netrin-1 expression is upregulated in MCL and Activated B-Cell DLBCL. On the opposit, DCC expression is downregulated in Germinal Centre DLBCL. From a therapeutical point of view, reinducing DCC's apoptotis by blocking its interaction with netrin-1 triggers tumor volum in a xenograft model. DCC-netrin-1 might also be involved in extra-nodal lymphoma development. Axe 2 : Functional caracterization of ADAMTSL1.DCC's ortholog in C. elegans, UNC-40, functionnaly interacts with the protein MADD-4 during the setting up of neuromuscular junctions. We did not find any such connexion between DCC and MADD-4's ortholog in mammals, ADAMTSL1. The function of ADAMTSL1 is unknown and we used different in vitro and in vivo models including a mouse model invalidated for this gene to unravel the function of this matricial protein, notably in muscular and cartilaginous tissus

Tese de Mestrado Integrado. Engenharia Informática e Computação. Faculdade de Engenharia. Universidade do Porto. 2012

This study tests whether contagion effects existed during the ¡§subprime mortgage crisis¡¨ among the equity markets of the US, the EU, Asia and emerging markets. The time-varying correlation coefficients are estimated by the dynamic conditional correlation (DCC) of Engle (2002), using a multivariate GJR-GARCH with AR (1) model. The empirical findings show that the conditional correlation coefficients of stock returns between the U.S. and others countries were positive and that the contagion effect exists among stock markets. Financial markets displayed contagion effects, in that the global equity markets were confronted with elevated systematic risk at the same time. Therefore, this study further examines the role of systematic risk in the equity market of each country. I used the rolling formulae, the MV-DGP, and DCC-GARCH (1, 1) models to estimate the CAPM beta and downside betas. This study found higher systematic risk (downside systematic risk) in the stock markets of the United States, Germany, France and Brazil, which had beta values nearly above one, while the Chinese stock market had the lowest systemic risk and served as a hedge for investors and fund managers. Finally, the results demonstrate that DCC-HW beta can capture some downside linkages between the market portfolios and expected stock returns, while these linkages cannot likely be captured by the CAPM beta.

The Dynamic Conditional Correlation (DCC) model proposed by Engle (2002) has become one of the most popular models for the analysis of multivariate financial time series. Yet, the impact of temporal aggregation on the DCC estimates has not yet been rigorously investigated. This thesis examines the changes of DCC estimates when the intraday returns are aggregated from 5-minutes to 270-minutes returns using Taiwanese eight industry index returns from Jan. 2, 2004 to Dec. 31, 2006. Our empirical analysis finds that dynamic correlation coefficients between the 8 industry index returns are all positive and time-varying. Further, Electronic and Building indices seem to have high correlation with other industry indices whereas plastics has a lower correlation with others. What is more important, all return series have higher conditional correlation for lower frequencies. In other words, temporary aggregation will increase the conditional correlation. This thesis also seeks to categorize the loan accounts of small- and medium-scale corporations according to their respective business sectors and calculate the monthly returns and standard deviation of the bank loans according to the groups of sample of credit records from each sector, with the purpose of establishing the efficient frontier of the loan combinations of the banks and estimation the dynamic conditional correlation to discover the optimal crediting policy. It is expected that the discussion using the model presented in the thesis may provide the basis for financial institutions as they establish their respective crediting policies.

Project and Report (M. Eng.)--Virginia Polytechnic Institute and State University, 1991. / Includes bibliographical references (leaf 236). Also available via the Internet.

This master thesis deals with the modeling of multivariate volatility in finan- cial time series. The aim of this work is to describe in detail selected approaches to modeling multivariate financial volatility, including verification of models, and then apply them in an empirical study of asset portfolio optimization. The results are compared with the classical approach of portfolio optimization theory based on unconditional moment estimates. The evaluation was based on four known op- timization problems, namely minimization of variance, Markowitz's model, ma- ximization of the Sharpe ratio and minimization of CVaR. The output portfolios were compared by using four metrics that reflect the returns and risks of the port- folios. The results demonstrated that employing the multivariate volatility models one obtains higher expected returns with less expected risk when comparing with the classical approach. 1