Neste trabalho propomos a utilização do método da função característica empírica (ECF - empirical characteristic function), para estimação do modelo de duração condicional estocástica (SCD - stochastic conditional duration). Para determinação das variáveis latentes do processo utilizamos três alternativas: um filtro de Kalman, um filtro obtido por integração numérica e um filtro baseado na expansão de Gram-Charlier até 4ª ordem. Os resultados são então aplicados em séries de duração da GE, Microsoft e USD/EUR. / We propose the use of the empirical characteristic function (ECF) method to estimate the parameters of the stochastic conditional duration (SCD) model. In order to estimate the latent variables we propose the use of three alternatives: a Kalman filter, a filter based on numerical integration (quadrature) and a filter based on the 4th-order Gram-Charlier expansion. The results are applied to the estimation of the parameters of the duration process for GE, Microsoft and USD/EUR.

Neste trabalho propomos a utilização do método da função característica empírica (ECF - empirical characteristic function), para estimação do modelo de duração condicional estocástica (SCD - stochastic conditional duration). Para determinação das variáveis latentes do processo utilizamos três alternativas: um filtro de Kalman, um filtro obtido por integração numérica e um filtro baseado na expansão de Gram-Charlier até 4ª ordem. Os resultados são então aplicados em séries de duração da GE, Microsoft e USD/EUR. / We propose the use of the empirical characteristic function (ECF) method to estimate the parameters of the stochastic conditional duration (SCD) model. In order to estimate the latent variables we propose the use of three alternatives: a Kalman filter, a filter based on numerical integration (quadrature) and a filter based on the 4th-order Gram-Charlier expansion. The results are applied to the estimation of the parameters of the duration process for GE, Microsoft and USD/EUR.

Thesis (PhD)--Macquarie University, Division of Economic and Financial Studies, Dept. of Statistics, 2008. / Bibliography: p. 73-77. / Introduction -- Lévy processes used in option pricing -- Option pricing for Lévy processes -- Option pricing based on empirical characteristic functions -- Performance of the five models on historical data -- Conclusions -- References -- Appendix A. Proofs -- Appendix B. Supplements -- Appendix C. Matlab programs. / Pricing problems of financial derivatives are among the most important ones in Quantitative Finance. Since 1973 when a Nobel prize winning model was introduced by Black, Merton and Scholes the Brownian Motion (BM) process gained huge attention of professionals professionals. It is now known, however, that stock market log-returns do not follow the very popular BM process. Derivative pricing models which are based on more general Lévy processes tend to perform better. --Carr & Madan (1999) and Lewis (2001) (CML) developed a method for vanilla options valuation based on a characteristic function of asset log-returns assuming that they follow a Lévy process. Assuming that at least part of the problem is in adequate modeling of the distribution of log-returns of the underlying price process, we use instead a nonparametric approach in the CML formula and replaced the unknown characteristic function with its empirical version, the Empirical Characteristic Functions (ECF). We consider four modifications of this model based on the ECF. The first modification requires only historical log-returns of the underlying price process. The other three modifications of the model need, in addition, a calibration based on historical option prices. We compare their performance based on the historical data of the DAX index and on ODAX options written on the index between the 1st of June 2006 and the 17th of May 2007. The resulting pricing errors show that one of our models performs, at least in the cases considered in the project, better than the Carr & Madan (1999) model based on calibration of a parametric Lévy model, called a VG model. --Our study seems to confirm a necessity of using implied parameters, apart from an adequate modeling of the probability distribution of the asset log-returns. It indicates that to precisely reproduce behaviour of the real option prices yet other factors like stochastic volatility need to be included in the option pricing model. Fortunately the discrepancies between our model and real option prices are reduced by introducing the implied parameters which seem to be easily modeled and forecasted using a mixture of regression and time series models. Such approach is computationaly less expensive than the explicit modeling of the stochastic volatility like in the Heston (1993) model and its modifications. / Mode of access: World Wide Web. / x, 111 p. ill., charts

Aim of this thesis is provide basic information about characteristic function used in statistic and compare its properties with the Fourier transform used in engineering applications. First part of this thesis is theoretical, there are discussed basic concepts, their properties and mutual relations. The second part is devoted to some possible applications, for example normality testing of data or utilization of the characteristic function in independent component analysis. The first chapter describes the introduction to probability theory for the unification of terminology and mentioned concepts will be used to demonstrate the interesting properties of characteristic function. The second chapter describes the Fourier transform, definition of characteristic function and their comparison. The second part of this text is devoted to applications the empirical characteristic function is analyzed as an estimate of the characteristic function of examined data. As an example of application is describe a simple test of normality. The last part deals with more advanced applications of characteristic function for methods such as independent component analysis.

Submitted by Haroudo Xavier Filho (haroudo.xavierfo@ufpe.br) on 2016-04-05T14:28:43Z No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) dissertacao_Raphaela(CD).pdf: 1067677 bytes, checksum: 6d371901336a7515911aeffd9ee38c74 (MD5) / Made available in DSpace on 2016-04-05T14:28:43Z (GMT). No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) dissertacao_Raphaela(CD).pdf: 1067677 bytes, checksum: 6d371901336a7515911aeffd9ee38c74 (MD5) Previous issue date: 2015-07-31 / CAPES / Este trabalho analisa propriedades da família de distribuições de probabilidade Composta N e propõe a sub-família Composta Poisson-Truncada como um meio de compor distribuições de probabilidade. Suas propriedades foram estudadas e uma nova distribuição foi investigada: a distribuição Composta Poisson-Truncada Normal. Esta distribuição possui três parâmetros e tem uma flexibilidade para modelar dados multimodais. Demonstramos que sua densidade é dada por uma mistura infinita de densidades normais em que os pesos são dados pela função de massa de probabilidade da Poisson-Truncada. Dentre as propriedades exploradas desta distribuição estão a função característica e expressões para o cálculo dos momentos. Foram analisados três métodos de estimação para os parâmetros da distribuição Composta Poisson-Truncada Normal, sendo eles, o método dos momentos, o da função característica empírica (FCE) e o método de máxima verossimilhança (MV) via algoritmo EM. Simulações comparando estes três métodos foram realizadas e, por fim, para ilustrar o potencial da distribuição proposta, resultados numéricos com modelagem de dados reais são apresentados. / This work analyzes properties of the Compound N family of probability distributions and proposes the sub-family Compound Poisson-Truncated as a means of composing probability distributions. Its properties were studied and a new distribution was investigated: the Compound Poisson-Truncated Normal distribution. This distribution has three parameters and has the flexibility to model multimodal data. We demonstrated that its density is given by an infinite mixture of normal densities where in the weights are given by the Poisson-Truncated probability mass function. Among the explored properties of this distribution are the characteristic function end expressions for the calculation of moments. Three estimation methods were analyzed for the parameters of the Compound Poisson-Truncated Normal distribution, namely, the method of moments, the empirical characteristic function (ECF) and the method of maximum likelihood (ML) by EM algorithm. Simulations comparing these three methods were performed and, finally, to illustrate the potential of the proposed distribution numerical results with real data modeling are presented.

Cette thèse est consacré aux tests d'adéquation basés sur la fonction caractéristique. Nous débutons en présentant et en complétant les résultats probabilistes nécessaires à la construction de statistiques de test prenant la fonction caractéristique et son pendant la fonction caractéristique empirique comme représentations respectives des lois de référence et de la loi inconnue de l'échantillon de vecteurs aléatoires à tester. Nous poursuivons le travail en faisant la revue et en classant les tests basés sur la fonction caractéristique existants. Nous élaborons ensuite une classe de statistiques de test s'appuyant sur le calcul d'une distance intégrale. Le cas de la distance L2 est étudié plus à fond, car nous avons pu établir des résultats asymptotiques dans ce dernier cas. Ceux-ci font intervenir les éléments propres inconnus d'un opérateur intégral. Nous présentons, puis utilisons, une méthode d'approximation spectrale basée sur une projection de l'opérateur sur une base orthonormée.Finalement, nous construisons une nouvelle classe de tests appartenant au paradigme des tests lisses de Neyman. L'étude précédente nous permet de simplifier considérablement la construction de ces tests, dont différentes versions sont proposées tant pour le test d'une hypothèse simple que pour le test d'une hypothèse composite. / This PhD thesis consists in building goodness-of-fit tests using the characteristic function (CF) as a prefered representation for the probability laws involved.We start with listing and improving results in probability theory necessary to build test statistics using the characteristic function and its conterpart the empirical characteristic function.We list and classify existing characteristic function based goodness-of-fit tests published by varions authors since 1977.Then, we build a class of tests based on integral metrics. We take particular attention to the case where the statistics are build using a L2 distance. More specifically, we give asymptotic results in this case. However, these results reveal the need for information on the unknown eigenelements of an integral operator. Thus, we present and implement an approximation method using a sequence of projections on orthonormal bases ofan hilbertian functional space.Finally, we will build another class of tests using the Neyman smooth test paradigm. This study is based on our previous results, that fit well into the construction of characteristic function based smooth tests. We give various applications, presenting tests for both a simple hypothesis and a composite hypothesis.