This PhD thesis aims to study financial processes which have semi-heavy-tailed marginal distributions and may exhibit memory. The traditional Black-Scholes model is expanded to incorporate memory via an integral operator, resulting in a class of market models which still preserve the completeness and arbitragefree conditions needed for replication of contingent claims. This approach is used to estimate the implied volatility of the resulting model. The first part of the thesis investigates the semi-heavy-tailed behaviour of financial processes. We treat these processes as continuous-time random walks characterised by a transition probability density governed by a fractional Riesz- Bessel equation. This equation extends the Feller fractional heat equation which generates a-stable processes. These latter processes have heavy tails, while those processes generated by the fractional Riesz-Bessel equation have semi-heavy tails, which are more suitable to model financial data. We propose a quasi-likelihood method to estimate the parameters of the fractional Riesz- Bessel equation based on the empirical characteristic function. The second part considers a dynamic model of complete financial markets in which the prices of European calls and puts are given by the Black-Scholes formula. The model has memory and can distinguish between historical volatility and implied volatility. A new method is then provided to estimate the implied volatility from the model. The third part of the thesis considers the problem of classification of financial markets using high-frequency data. The classification is based on the measure representation of high-frequency data, which is then modelled as a recurrent iterated function system. The new methodology developed is applied to some stock prices, stock indices, foreign exchange rates and other financial time series of some major markets. In particular, the models and techniques are used to analyse the SET index, the SET50 index and the MAI index of the Stock Exchange of Thailand.

Financial processes may possess long memory and their probability densities may display heavy tails. Many models have been developed to deal with this tail behaviour, which reflects the jumps in the sample paths. On the other hand, the presence of long memory, which contradicts the efficient market hypothesis, is still an issue for further debates. These difficulties present challenges with the problems of memory detection and modelling the co-presence of long memory and heavy tails. This PhD project aims to respond to these challenges. The first part aims to detect memory in a large number of financial time series on stock prices and exchange rates using their scaling properties. Since financial time series often exhibit stochastic trends, a common form of nonstationarity, strong trends in the data can lead to false detection of memory. We will take advantage of a technique known as multifractal detrended fluctuation analysis (MF-DFA) that can systematically eliminate trends of different orders. This method is based on the identification of scaling of the q-th-order moments and is a generalisation of the standard detrended fluctuation analysis (DFA) which uses only the second moment; that is, q = 2. We also consider the rescaled range R/S analysis and the periodogram method to detect memory in financial time series and compare their results with the MF-DFA. An interesting finding is that short memory is detected for stock prices of the American Stock Exchange (AMEX) and long memory is found present in the time series of two exchange rates, namely the French franc and the Deutsche mark. Electricity price series of the five states of Australia are also found to possess long memory. For these electricity price series, heavy tails are also pronounced in their probability densities. The second part of the thesis develops models to represent short-memory and longmemory financial processes as detected in Part I. These models take the form of continuous-time AR(∞) -type equations whose kernel is the Laplace transform of a finite Borel measure. By imposing appropriate conditions on this measure, short memory or long memory in the dynamics of the solution will result. A specific form of the models, which has a good MA(∞) -type representation, is presented for the short memory case. Parameter estimation of this type of models is performed via least squares, and the models are applied to the stock prices in the AMEX, which have been established in Part I to possess short memory. By selecting the kernel in the continuous-time AR(∞) -type equations to have the form of Riemann-Liouville fractional derivative, we obtain a fractional stochastic differential equation driven by Brownian motion. This type of equations is used to represent financial processes with long memory, whose dynamics is described by the fractional derivative in the equation. These models are estimated via quasi-likelihood, namely via a continuoustime version of the Gauss-Whittle method. The models are applied to the exchange rates and the electricity prices of Part I with the aim of confirming their possible long-range dependence established by MF-DFA. The third part of the thesis provides an application of the results established in Parts I and II to characterise and classify financial markets. We will pay attention to the New York Stock Exchange (NYSE), the American Stock Exchange (AMEX), the NASDAQ Stock Exchange (NASDAQ) and the Toronto Stock Exchange (TSX). The parameters from MF-DFA and those of the short-memory AR(∞) -type models will be employed in this classification. We propose the Fisher discriminant algorithm to find a classifier in the two and three-dimensional spaces of data sets and then provide cross-validation to verify discriminant accuracies. This classification is useful for understanding and predicting the behaviour of different processes within the same market. The fourth part of the thesis investigates the heavy-tailed behaviour of financial processes which may also possess long memory. We consider fractional stochastic differential equations driven by stable noise to model financial processes such as electricity prices. The long memory of electricity prices is represented by a fractional derivative, while the stable noise input models their non-Gaussianity via the tails of their probability density. A method using the empirical densities and MF-DFA will be provided to estimate all the parameters of the model and simulate sample paths of the equation. The method is then applied to analyse daily spot prices for five states of Australia. Comparison with the results obtained from the R/S analysis, periodogram method and MF-DFA are provided. The results from fractional SDEs agree with those from MF-DFA, which are based on multifractal scaling, while those from the periodograms, which are based on the second order, seem to underestimate the long memory dynamics of the process. This highlights the need and usefulness of fractal methods in modelling non-Gaussian financial processes with long memory.

Le travail réalisé dans le cadre de cette thèse a pour but de caractériser des signaux aléatoires, rencontrés dans le domaine aérien et sous-marin, en s’appuyant sur une approche statistique. En traitement du signal, l'analyse statistique a longtemps été fondée sous l'hypothèse de Gaussianité des données. Cependant, ce modèle n'est plus valide dès lors que la densité de probabilité des données se caractérise par des phénomènes de queues lourdes et d'asymétrie. Une famille de lois est particulièrement adaptée pour représenter de tels phénomènes : les distributions α-stables. Dans un premier temps, les distributions α-stables ont été présentées et utilisées pour estimer des données synthétiques et réelles, issues d'un sondeur monofaisceau, dans une stratégie de classification de fonds marins. La classification est réalisée à partir de la théorie des fonctions de croyance, permettant ainsi de prendre en compte l'imprécision et l'incertitude liées aux données et à l'estimation de celles-ci. Les résultats obtenus ont été comparés à un classifieur Bayésien. Dans un second temps, dans le contexte de la surveillance maritime, une approche statistique à partir des distributions α-stables a été réalisée afin de caractériser les échos indésirables réfléchis par la surface maritime, appelés aussi fouillis de mer, où la surface est observée en configuration bistatique. La surface maritime a d'abord été générée à partir du spectre d'Elfouhaily puis la Surface Équivalente Radar (SER) de celle-ci a été déterminée à partir de l'Optique Physique (OP). Les distributions de Weibull et ont été utilisées et comparées au modèle α-stable. La validité de chaque modèle a été étudiée à partir d'un test de Kolmogorov-Smirnov. / The purpose of this thesis is to characterize random signals, observed in air and underwater context, by using a statistical approach. In signal processing, the hypothesis of Gaussian model is often used for a statistical study. However, the Gaussian model is not valid when the probability density function of data are characterized by heavy-tailed and skewness phenomena. A family of laws can fit these phenomena: the α-stable distributions. Firstly, the α-stable distribution have been used to estimate generated and real data, extracted from a mono-beam echo-sounder, for seabed sediments classification. The classification is made by using the theory of belief functions, which can take into account the imprecision and uncertainty of data and theirs estimations. The results have been compared to a Bayesian approach. Secondly, in a context a marine surveillance, a statistical study from the α-stable distribution has been made to characterize undesirable echo reflected by a sea surface, called sea clutter, where the sea surface is considered in a bistatic configuration. The sea surface has been firstly generated by the Elfouhaily sea spectrum and the Radar Cross Section (RCS) of the sea surface has been computed by the Physical Optics (PO). The Weibull and distributions have been used and the results compared to the α-stable model. The validity of each model has been evaluated by a Kolmogorov-Smirnov test.

La première partie de cette thèse introduit une nouvelle méthodologie pour la réalisation d’exercices de stress-tests. Notre approche permet de considérer des scénarios de stress beaucoup plus riches qu’en pratique, qui évaluent l’impact d’une modification de la distribution statistique des facteurs influençant les prix d’actifs, pas uniquement les conséquences d’une réalisation particulière de ces facteurs, et prennent en compte la réaction du gestionnaire de portefeuille au choc. La deuxième partie de la thèse est consacrée à la valorisation des obligations à maturité très longues (supérieure à 10 ans). La modélisation de la volatilité des taux de très long terme est un défi, notamment du fait des contraintes posées par l’absence d’opportunités d’arbitrage, et la plupart des modèles de taux d’intérêt en absence d’opportunités d’arbitrage impliquent un taux limite (de maturité infinie) constant. Le deuxième chapitre étudie la compatibilité du facteur "niveau", dont les variations ont un impact uniforme sur l’ensemble des taux modélisés, a fortiori les plus longs, avec l’absence d’opportunités d’arbitrage. Nous introduisons dans le troisième chapitre une nouvelle classe de modèle de taux d’intérêt, sans opportunités d’arbitrage, où le taux limite est stochastique, dont nous présentons les propriétés empiriques sur une base de données de prix d’obligations du Trésor américain. / In the first part of this thesis, we introduce a new methodology for stress-test exercises. Our approach allows to consider richer stress-test exercises, which assess the impact of a modification of the whole distribution of asset prices’ factors, rather than focusing as the common practices on a single realization of these factors, and take into account the potential reaction to the shock of the portfolio manager.
 The second part of the thesis is devoted to the pricing of bonds with very long-term time-to-maturity (more than ten years). Modeling the volatility of very long-term rates is a challenge, due to the constraints put by no-arbitrage assumption. As a consequence, most of the no-arbitrage term structure models assume a constant limiting rate (of infinite maturity). The second chapter investigates the compatibility of the so-called "level" factor, whose variations have a uniform impact on the modeled yield curve, with the no-arbitrage assumptions. We introduce in the third chapter a new class of arbitrage-free term structure factor models, which allows the limiting rate to be stochastic, and present its empirical properties on a dataset of US T-Bonds.