No description available.

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.

For financial forcasting of crisis new concepts from disciplines dissimilar to economics are looked for by financial experts. The branch of econophysics using theories of natural sciences is significant. The meaning of this work is to point out one of many methods applied to financial data with help of the theory of turbulence of fluids and deterministic chaos. We provide a parallel analysis of high frequency financial time series of a stock index and velocities of a turbulent fluid. This work concerns the use of concepts from statistical mathematics, probability theory and scaling. We find differences of both studied systems but the methodologies of natural diciplines can be also applied to financial data.

This thesis investigates a new method for financial data simulation using novel neural network models developed by the author. Using two improved models for financial data simulation and prediction, the trigonometric polynomial higher order neural network group models have been developed. The theoretical principles of these improved models are presented and demonstrated in the thesis. It is the first attempt to use trigonometric polynomial high order neural network group models for financial data simulation. We could not find any references to using trigonometric polynomial high order neural network group models for financial data simulation in the extensive literature search conducted for this thesis, including a thorough Internet search on this topic. The author has developed a computer program, called 'THONG'. The program, running on X-windows, is based on the new neural network models developed, and also uses group models. This program allows users to apply in practice his new analysis and prediction method. The 'THONG' program is a user-friendly GUI system. All the steps of the operation in this system are easily controlled using a mouse. Both system operation and system mode can be viewed during the processing of data. THONG models have proven capable of handling high frequency, high order nonlinear and discontinuous data. The results of processing the experimental data using the THONG financial simulator are presented in the thesis. These results confirm that the THONG group models converge without difficulty, and are considerably more accurate than traditional neural network models. / Doctor of Philosophy (PhD)

Today, when businesses and organizations are generating huge volumes of data; the applications like Web 2.0 or social networking requires processing of petabytes of data. Stock Exchange Systems are among the ones that process large amount of quotes and trades on a daily basis. The limited database storage ability is a major bottleneck in meeting up the challenge of providing efficient access to information. Further to this, varying data are the major source of information for the financial industry. This data needs to be read and written efficiently in the database; this is quite costly when it comes to traditional Relational Database Management System. RDBMS is good for different scenarios and can handle certain types of data very well, but it isn’t always the perfect choice. The existence of innovative architectures allows the storage of large data in an efficient manner. “Not only SQL” brings an effective solution through the provision of an efficient information storage capability. NOSQL is an umbrella term for various new data store. The NOSQL databases have gained popularity due to different factors that include their open source nature, existence of non-relational data store, high-performance, fault-tolerance, and scalability to name a few. Nowadays, NOSQL databases are rapidly gaining popularity because of the advantages that they offer compared to RDBMS. The major aim of this research is to find an efficient solution for storing and processing the huge volume of data for certain variants. The study is based on choosing a reliable, distributed, and efficient NOSQL database at Cinnober Financial Technology AB. The research majorly explores NOSQL databases and discusses issues with RDBMS; eventually selecting a database, which is best suited for financial data management. It is an attempt to contribute the current research in the field of NOSQL databases which compares one such NOSQL database Apache Cassandra with Apache Lucene and the traditional relational database MySQL for financial management. The main focus is to find out which database is the preferred choice for different variants. In this regard, the performance test framework for a selected set of candidates has also been taken into consideration.

Cluster analysis, the automatic procedure by which large data sets can be split into similar groups of objects (clusters), has innumerable applications in a wide range of problem domains. Improvements in clustering quality (as captured by internal validation indexes) and speed (number of iterations until cost function convergence), the main focus of this work, have many desirable consequences. They can result, for example, in faster and more precise detection of illness onset based on symptoms or it could provide investors with a rapid detection and visualization of patterns in financial time series and so on. Partitional clustering, one of the most popular ways of doing cluster analysis, can be classified into two main categories: hard (where the clusters discovered are disjoint) and soft (also known as fuzzy; clusters are non-disjoint, or overlapping). In this work we consider how improvements in the speed and solution quality of the soft partitional clustering algorithm Fuzzy C-means (FCM) can be achieved through more careful and informed initialization based on data content. By carefully selecting the cluster centers in a way which disperses the initial cluster centers through the data space, the resulting FCM++ approach samples starting cluster centers during the initialization phase. The cluster centers are well spread in the input space, resulting in both faster convergence times and higher quality solutions. Moreover, we allow the user to specify a parameter indicating how far and apart the cluster centers should be picked in the dataspace right at the beginning of the clustering procedure. We show FCM++'s superior behaviour in both convergence times and quality compared with existing methods, on a wide rangeof artificially generated and real data sets. We consider a case study where we propose a methodology based on FCM++for pattern discovery on synthetic and real world time series data. We discuss a method to utilize both Pearson correlation and Multi-Dimensional Scaling in order to reduce data dimensionality, remove noise and make the dataset easier to interpret and analyse. We show that by using FCM++ we can make an positive impact on the quality (with the Xie Beni index being lower in nine out of ten cases for FCM++) and speed (with on average 6.3 iterations compared with 22.6 iterations) when trying to cluster these lower dimensional, noise reduced, representations of the time series. This methodology provides a clearer picture of the cluster analysis results and helps in detecting similarly behaving time series which could otherwise come from any domain. Further, we investigate the use of Spherical Fuzzy C-Means (SFCM) with the seeding mechanism used for FCM++ on news text data retrieved from a popular British newspaper. The methodology allows us to visualize and group hundreds of news articles based on the topics discussed within. The positive impact made by SFCM++ translates into a faster process (with on average 12.2 iterations compared with the 16.8 needed by the standard SFCM) and a higher quality solution (with the Xie Beni being lower for SFCM++ in seven out of every ten runs).

O aumento dos fluxos globais comerciais e financeiros, a partir da década de 90, e as diversas crises ocorridas até o atual período fizeram da avaliação de contágio um tema extremamente relevante, tanto para investidores quanto para formuladores de política. Nesse sentido, a presente dissertação tem como objetivo testar a hipótese de contágio financeiro para os mercados de Brasil, Inglaterra e Espanha em face à última crise americana de 2008. Para tanto, desenvolveu-se o artigo que integra o Capítulo 2 - a espinha dorsal deste trabalho - com dados diários dos retornos dos índices de Jan/2004 a Jun/2011. No âmbito da metodologia de cópulas, adotou-se uma estratégia empírica com base em duas etapas: i) a estimativa não paramétrica de cópulas, via kernel, utilizando o método desenvolvido em Fermanian et al. (2002) e a avaliação através de uma abordagem de bootstrap, sobre a ocorrência de um aumento significativo nas medidas de dependência delas extraídas; ii) testes sobre a igualdade entre cópulas empíricas, conforme proposto por Remillard e Scaillet (2009), a fim de verificar se houve mudança na estrutura de dependência a partir da crise. Os resultados obtidos nas duas etapas da estratégia empírica são semelhantes e sugerem a existência de contágio financeiro para os países analisados no período estudado. / The increase in global trade and financial flows since the 90’s, and the various crises in the current period until these days made contagion an extremely important issue for both investors and policy makers. Accordingly, this dissertation aims to test the hypothesis of financial contagion between USA and markets in Brazil, England and Spain in the face of the last USA crisis of 2008. To this end, we produce the article in Chapter 2 - the backbone of this work - with daily data of index-returns from Jan/2004 to Jun/2011. Under the scope of copula methodology, we addopt an empirical strategy based on two steps: i) estimating nonparametric copulas via kernel, using the method developed in Fermanian et al. (2002) and assessing through a bootstrap approach whether a significant change in dependence measures extracts thereof, ii) testing whether two empirical estimated copulas are the same, as proposed by Remillard e Scaillet (2009), to check again whether dependence structures change with crisis. The results obtained in these two steps of the empirical strategy are similar and suggest the existence of financial contagion between the countries analysed in the studied period.

O aumento dos fluxos globais comerciais e financeiros, a partir da década de 90, e as diversas crises ocorridas até o atual período fizeram da avaliação de contágio um tema extremamente relevante, tanto para investidores quanto para formuladores de política. Nesse sentido, a presente dissertação tem como objetivo testar a hipótese de contágio financeiro para os mercados de Brasil, Inglaterra e Espanha em face à última crise americana de 2008. Para tanto, desenvolveu-se o artigo que integra o Capítulo 2 - a espinha dorsal deste trabalho - com dados diários dos retornos dos índices de Jan/2004 a Jun/2011. No âmbito da metodologia de cópulas, adotou-se uma estratégia empírica com base em duas etapas: i) a estimativa não paramétrica de cópulas, via kernel, utilizando o método desenvolvido em Fermanian et al. (2002) e a avaliação através de uma abordagem de bootstrap, sobre a ocorrência de um aumento significativo nas medidas de dependência delas extraídas; ii) testes sobre a igualdade entre cópulas empíricas, conforme proposto por Remillard e Scaillet (2009), a fim de verificar se houve mudança na estrutura de dependência a partir da crise. Os resultados obtidos nas duas etapas da estratégia empírica são semelhantes e sugerem a existência de contágio financeiro para os países analisados no período estudado. / The increase in global trade and financial flows since the 90’s, and the various crises in the current period until these days made contagion an extremely important issue for both investors and policy makers. Accordingly, this dissertation aims to test the hypothesis of financial contagion between USA and markets in Brazil, England and Spain in the face of the last USA crisis of 2008. To this end, we produce the article in Chapter 2 - the backbone of this work - with daily data of index-returns from Jan/2004 to Jun/2011. Under the scope of copula methodology, we addopt an empirical strategy based on two steps: i) estimating nonparametric copulas via kernel, using the method developed in Fermanian et al. (2002) and assessing through a bootstrap approach whether a significant change in dependence measures extracts thereof, ii) testing whether two empirical estimated copulas are the same, as proposed by Remillard e Scaillet (2009), to check again whether dependence structures change with crisis. The results obtained in these two steps of the empirical strategy are similar and suggest the existence of financial contagion between the countries analysed in the studied period.

Le but de cette thèse est d’approfondir les connaissances académiques sur les variations jointes des actifs financiers à haute fréquence en les analysant sous un point de vue novateur. Nous tirons profit d’une base de données de prix tick-by-tick pour mettre en lumière de nouveaux faits stylises sur la corrélation haute fréquence, et également pour tester la validité empirique de modèles multivariés. Dans le chapitre 1, nous discutons des raisons pour lesquelles la corrélation haute fréquence est d’une importance capitale pour le trading. Par ailleurs, nous passons en revue la littérature empirique et théorique sur la corrélation à de petites échelles de temps. Puis nous décrivons les principales caractéristiques du jeu de données que nous utilisons. Enfin, nous énonçons les résultats obtenus dans cette thèse. Dans le chapitre 2, nous proposons une extension du modèle de subordination au cas multivarié. Elle repose sur la définition d’un temps événementiel global qui agrège l’activité financière de tous les actifs considérés. Nous testons la capacité de notre modèle à capturer les propriétés notables de la distribution multivariée empirique des rendements et observons de convaincantes similarités. Dans le chapitre 3, nous étudions les relations lead/lag à haute fréquence en utilisant un estimateur de fonction de corrélation adapte aux données tick-by-tick. Nous illustrons sa supériorité par rapport à l’estimateur standard de corrélation pour détecter le phénomène de lead/lag. Nous établissons un parallèle entre le lead/lag et des mesures classiques de liquidité et révélons un arbitrage pour déterminer les paires optimales pour le trading de lead/lag. Enfin, nous évaluons la performance d’un indicateur basé sur le lead/lag pour prévoir l’évolution des prix à court terme. Dans le chapitre 4, nous nous intéressons au profil saisonnier de la corrélation intra-journalière. Nous estimons ce profil sur quatre univers d’actions et observons des ressemblances frappantes. Nous tentons d’incorporer ce fait stylise dans un modèle de prix tick-by-tick base sur des processus de Hawkes. Le modèle ainsi construit capture le profil de corrélation empirique assez finement, malgré sa difficulté à atteindre le niveau de corrélation absolu. / This thesis aims at providing insight into comovements of financial assets at high frequency from an original point of view. We take advantage of a database of tick-by-tick prices to bring to light new stylized facts on high frequency correlation as well as to check the empirical validity of multivariate modelling frameworks. In chapter 1, we elaborate on the reasons why high frequency correlation is of the utmost importance for trading purposes. We also briefly review the empirical and theoretical literature on correlation at small time scales. Then, we describe the main features of the data set we use. Finally, we enunciate the results obtained in this thesis. In chapter 2, we suggest a way of extending the subordination modelling to the multivariate case. This relies on the definition of a global event time that merges the trading activity of all assets under consideration. We test the ability of our model to capture salient features of the empirical multivariate probability distribution of returns and find a convincing agreement. In chapter 3, we study high frequency lead/lag relationships using a suitable cross-correlation estimator for tick-by-tick data. We show its superiority over the classical correlation estimator in detecting lead/lag patterns. We relate lead/lag to standard liquidity measures and exhibit a trade-off to find optimal pairs for lead/lag trading. Finally, we evaluate the performance of a lead/lag indicator in forecasting the short-term evolution of prices. In chapter 4, we focus on the intraday correlation seasonal pattern. We estimate this pattern over four universes of stocks and observe striking similarities. We attempt to incorporate this stylized fact into a tick-by-tick price model based upon Hawkes processes. The resulting model captures the empirical profile of correlation quite well, though it doesn’t match the absolute level of correlation.

O aumento dos fluxos globais comerciais e financeiros, a partir da década de 90, e as diversas crises ocorridas até o atual período fizeram da avaliação de contágio um tema extremamente relevante, tanto para investidores quanto para formuladores de política. Nesse sentido, a presente dissertação tem como objetivo testar a hipótese de contágio financeiro para os mercados de Brasil, Inglaterra e Espanha em face à última crise americana de 2008. Para tanto, desenvolveu-se o artigo que integra o Capítulo 2 - a espinha dorsal deste trabalho - com dados diários dos retornos dos índices de Jan/2004 a Jun/2011. No âmbito da metodologia de cópulas, adotou-se uma estratégia empírica com base em duas etapas: i) a estimativa não paramétrica de cópulas, via kernel, utilizando o método desenvolvido em Fermanian et al. (2002) e a avaliação através de uma abordagem de bootstrap, sobre a ocorrência de um aumento significativo nas medidas de dependência delas extraídas; ii) testes sobre a igualdade entre cópulas empíricas, conforme proposto por Remillard e Scaillet (2009), a fim de verificar se houve mudança na estrutura de dependência a partir da crise. Os resultados obtidos nas duas etapas da estratégia empírica são semelhantes e sugerem a existência de contágio financeiro para os países analisados no período estudado. / The increase in global trade and financial flows since the 90’s, and the various crises in the current period until these days made contagion an extremely important issue for both investors and policy makers. Accordingly, this dissertation aims to test the hypothesis of financial contagion between USA and markets in Brazil, England and Spain in the face of the last USA crisis of 2008. To this end, we produce the article in Chapter 2 - the backbone of this work - with daily data of index-returns from Jan/2004 to Jun/2011. Under the scope of copula methodology, we addopt an empirical strategy based on two steps: i) estimating nonparametric copulas via kernel, using the method developed in Fermanian et al. (2002) and assessing through a bootstrap approach whether a significant change in dependence measures extracts thereof, ii) testing whether two empirical estimated copulas are the same, as proposed by Remillard e Scaillet (2009), to check again whether dependence structures change with crisis. The results obtained in these two steps of the empirical strategy are similar and suggest the existence of financial contagion between the countries analysed in the studied period.