Pagrindinė šios disertacijos tema - stochastinių diferencialinių lygčių (SDL), valdomų trupmeninio Brauno judesio (tBj), sprendinių Hursto indekso H vertinimas. Pirmiausia disertacijoje išnagrinėta SDL, valdomų tBj, sprendinių pirmos ir antros eilės kvadratinių variacijų ribinė elgsena. Iš šių rezultatų seka keli stipriai pagrįsti Hursto indekso H įvertiniai. Įrodyta, kad šie įvertiniai išlieka stipriai pagrįsti, jei tikra sprendinio trajektorija keičiama jos Milšteino aproksimacija. Taip pat išnagrinėtos pokyčių santykio (increment ratios) statistikos H įvertinio, gauto J. M. Bardeto ir D. Surgailio 2010 m., taikymo trupmeninio geometrinio Brauno judesio Hursto indekso vertinimui galimybės bei nustatytas modifikuoto Gladyševo H įvertinio konvergavimo į tikrąją parametro reikšmę greitis. Gauti įvertiniai palyginti su kai kuriais kitais žinomais Hursto indekso H įvertiniais: naiviais bei mažiausių kvadratų Gladyševo ir eta-sumavimo osciliacijos įvertiniais, variogramos įvertiniu ir pokyčių santykio statistikos įvertiniu. Įvertiniu elgsena buvo palyginta trupmeniniam Ornšteino-Ulenbeko (OU) procesui bei trupmeniniam geometriniam Brauno judesiui (gBj). Pradinės išvados buvo padarytos O-U procesui, kuris yra Gauso, o gBj procesas buvo naudojamas patikrinti, kaip šie įvertiniai elgiasi, kai procesas yra ne Gauso. Disertaciją sudaro įvadas, 3 pagrindiniai skyriai, išvados, literatūros sąrašas, autoriaus publikacijų disertacijos tema sąrašas ir du priedai. / The main topic of this dissertation is the estimation of the Hurst index H of the solutions of stochastic differential equations (SDEs) driven by the fractional Brownian motion (fBm). Firstly, the limit behavior of the first and second order quadratic variations of the solutions of SDEs driven by the fBm is analyzed. This yields several strongly consistent estimators of the Hurst index H. Secondly, it is proved that in case the solution of the SDE is replaced by its Milstein approximation, the estimators remain strongly consistent. Additionally, the possibilities of applying the increment ratios (IR) statistic based estimator of H originally obtained by J. M. Bardet and D. Surgailis in 2010 to the fractional geometric Brownian motion are examined. Furthermore, this dissertation derives the convergence rate of the modified Gladyshev's estimator of the Hurst index to its real value. The estimators obtained in the dissertation were compared with several other known estimators of the Hurst index H, namely the naive and ordinary least squares Gladyshev and eta-summing oscillation estimators, the variogram estimator and the IR estimator. The models chosen for comparison of these estimators were the fractional Ornstein-Uhlenbeck (O-U) process and the fractional geometric Brownian motion (gBm). The initial inference about the behavior of these estimators was drawn for the O-U process which is Gaussian, while the gBm process was used to check how the estimators behave in a... [to full text]

The main topic of this dissertation is the estimation of the Hurst index H of the solutions of stochastic differential equations (SDEs) driven by the fractional Brownian motion (fBm). Firstly, the limit behavior of the first and second order quadratic variations of the solutions of SDEs driven by the fBm is analyzed. This yields several strongly consistent estimators of the Hurst index H. Secondly, it is proved that in case the solution of the SDE is replaced by its Milstein approximation, the estimators remain strongly consistent. Additionally, the possibilities of applying the increment ratios (IR) statistic based estimator of H originally obtained by J. M. Bardet and D. Surgailis in 2010 to the fractional geometric Brownian motion are examined. Furthermore, this dissertation derives the convergence rate of the modified Gladyshev’s estimator of the Hurst index to its real value. The estimators obtained in the dissertation were compared with several other known estimators of the Hurst index H, namely the naive and ordinary least squares Gladyshev and eta-summing oscillation estimators, the variogram estimator and the IR estimator. The models chosen for comparison of these estimators were the fractional Ornstein-Uhlenbeck (O-U) process and the fractional geometric Brownian motion (gBm). The initial inference about the behavior of these estimators was drawn for the O-U process which is Gaussian, while the gBm process was used to check how the estimators behave in a... [to full text] / Pagrindinė šios disertacijos tema – stochastinių diferencialinių lygčių (SDL), valdomų trupmeninio Brauno judesio (tBj), sprendinių Hursto indekso H vertinimas. Pirmiausia disertacijoje išnagrinėta SDL, valdomų tBj, sprendinių pirmos ir antros eilės kvadratinių variacijų ribinė elgsena. Iš šių rezultatų seka keli stipriai pagrįsti Hursto indekso H įvertiniai. Įrodyta, kad šie įvertiniai išlieka stipriai pagrįsti, jei tikra sprendinio trajektorija keičiama jos Milšteino aproksimacija. Taip pat išnagrinėtos pokyčių santykio (increment ratios) statistikos H įvertinio, gauto J. M. Bardeto ir D. Surgailio 2010 m., taikymo trupmeninio geometrinio Brauno judesio Hursto indekso vertinimui galimybės bei nustatytas modifikuoto Gladyševo H įvertinio konvergavimo i tikrąją parametro reikšme greitis. Gauti įvertiniai palyginti su kai kuriais kitais žinomais Hursto indekso H įvertiniais: naiviais bei mažiausių kvadratų Gladyševo ir eta-sumavimo osciliacijos įvertiniais, variogramos įvertiniu ir pokyčių santykio statistikos įvertiniu. Įvertinių elgsena buvo palyginta trupmeniniam Ornšteino-Ulenbeko (OU) procesui bei trupmeniniam geometriniam Brauno judesiui (gBj). Pradinės išvados buvo padarytos O-U procesui, kuris yra Gauso, o gBj procesas buvo naudojamas patikrinti, kaip šie įvertiniai elgiasi, kai procesas yra ne Gauso. Disertaciją sudaro įvadas, 3 pagrindiniai skyriai, išvados, literatūros sąrašas, autoriaus publikacijų disertacijos tema sąrašas ir du priedai.

This thesis focuses on two statistical problems related to credit scoring. In credit scoring of individuals, two classes are distinguished, namely low and high risk individuals (the so-called "good" and "bad" risk classes). Firstly, we suggest a measure which may be used to study the nature of a classifier for distinguishing between the two risk classes. Secondly, we derive a new method DOUW (detecting outliers using weights) which may be used to fit logistic regression models robustly and for the detection of outliers. In the first problem, the focus is on a measure which may be used to study the nature of a classifier. This measure transforms a random variable so that it has the same distribution as another random variable. Assuming a linear form of this measure, three methods for estimating the parameters (slope and intercept) and for constructing confidence bands are developed and compared by means of a Monte Carlo study. The application of these estimators is illustrated on a number of datasets. We also construct statistical hypothesis to test this linearity assumption. In the second problem, the focus is on providing a robust logistic regression fit and the identification of outliers. It is well-known that maximum likelihood estimators of logistic regression parameters are adversely affected by outliers. We propose a robust approach that also serves as an outlier detection procedure and is called DOUW. The approach is based on associating high and low weights with the observations as a result of the likelihood maximization. It turns out that the outliers are those observations to which low weights are assigned. This procedure depends on two tuning constants. A simulation study is presented to show the effects of these constants on the performance of the proposed methodology. The results are presented in terms of four benchmark datasets as well as a large new dataset from the application area of retail marketing campaign analysis. In the last chapter we apply the techniques developed in this thesis on a practical credit scoring dataset. We show that the DOUW method improves the classifier performance and that the measure developed to study the nature of a classifier is useful in a credit scoring context and may be used for assessing whether the distribution of the good and the bad risk individuals is from the same translation-scale family. / Thesis (Ph.D. (Risk Analysis))--North-West University, Potchefstroom Campus, 2008.

This thesis focuses on two statistical problems related to credit scoring. In credit scoring of individuals, two classes are distinguished, namely low and high risk individuals (the so-called "good" and "bad" risk classes). Firstly, we suggest a measure which may be used to study the nature of a classifier for distinguishing between the two risk classes. Secondly, we derive a new method DOUW (detecting outliers using weights) which may be used to fit logistic regression models robustly and for the detection of outliers. In the first problem, the focus is on a measure which may be used to study the nature of a classifier. This measure transforms a random variable so that it has the same distribution as another random variable. Assuming a linear form of this measure, three methods for estimating the parameters (slope and intercept) and for constructing confidence bands are developed and compared by means of a Monte Carlo study. The application of these estimators is illustrated on a number of datasets. We also construct statistical hypothesis to test this linearity assumption. In the second problem, the focus is on providing a robust logistic regression fit and the identification of outliers. It is well-known that maximum likelihood estimators of logistic regression parameters are adversely affected by outliers. We propose a robust approach that also serves as an outlier detection procedure and is called DOUW. The approach is based on associating high and low weights with the observations as a result of the likelihood maximization. It turns out that the outliers are those observations to which low weights are assigned. This procedure depends on two tuning constants. A simulation study is presented to show the effects of these constants on the performance of the proposed methodology. The results are presented in terms of four benchmark datasets as well as a large new dataset from the application area of retail marketing campaign analysis. In the last chapter we apply the techniques developed in this thesis on a practical credit scoring dataset. We show that the DOUW method improves the classifier performance and that the measure developed to study the nature of a classifier is useful in a credit scoring context and may be used for assessing whether the distribution of the good and the bad risk individuals is from the same translation-scale family. / Thesis (Ph.D. (Risk Analysis))--North-West University, Potchefstroom Campus, 2008.

The gravity model for international trade is one of the most successful empirical models in trade literature. There is a long tradition to log-linearise the multiplicative model and to estimate the parameters of interest by least squares. But this practice is inappropriate for several reasons. First of all, bilateral trade flows are frequently zero and disregarding countries that do not trade with each other produces biased results. Second, log-linearisation in the presence of heteroscedasticity leads to inconsistent estimates in general. In recent years, the Poisson gravity model along with pseudo maximum likelihood estimation methods have become popular as a way of dealing with such econometric issues as arise when dealing with origin-destination flows. But the standard Poisson model specification is vulnerable to problems of overdispersion and excess zero flows. To overcome these problems, this paper presents zero-inflated extensions of the Poisson and negative binomial specifications as viable alternatives to both the log-linear and the standard Poisson specifications of the gravity model. The performance of the alternative model specifications is assessed on a real world example, where more than half of country-level trade flows are zero. (authors' abstract) / Series: Working Papers in Regional Science

Uma questão interessante na política monetária é se os Bancos Centrais dão pesos iguais para desvios positivos e negativos da inflação e do hiato do produto das suas respectivas metas. Para responder à esta questão, estimou-se a função perda da autoridade monetária não parametricamente através do método de sieve estimator, expandindo-a através de uma base composta de polinômios ortogonais. A economia foi modelada com agentes foward-looking e com comprometimento por parte da autoridade monetária. O método foi aplicado para a os Estados Unidos desde 1960 e para o Brasil a partir de 1999. Para a economia norte-americana não foram encontradas evidências de assimetria nas preferências da autoridade monetária. Já no Brasil o Banco Central mostrou ter preferências assimétricas quanto à inflação, auferindo uma maior perda de desvios negativos do que positivos em relação à meta. / An interesting question in monetary policy is whether the Central Bank gives equal weights to positive and negative deviations of inflation and output gap from their targets. Trying answering this question, we estimated the monetary authority’s loss function nonparametrically, using the method of sieves, expanding it with orthogonal polynomials. The economy was model with forward-looking agents and with commitment of the monetary authority. We applied the method to U.S. monetary policy since 1960 and for Brazil since 1999. For the U.S. economy, it was not found evidence of asymmetry in the preferences of the monetary authority. In Brazil, the Central Bank proved to have asymmetric preferences about inflation, with a greater loss for negative deviations of inflation from the target rather for positive ones.

This work introduces a time adaptive strategy that uses a refinement estimator based on the geometric indicator of curvature. The developed methodology is suitable for problems of numerical time integration present, for example, in the study of bodies subjected to dynamical loads. The refinement estimator demands low computational resources, being easily applied to several direct integration methods. Trying to interact these methods, an object oriented library that uses the developed scheme of adaptation is built. The main idea of this tool is to incorporate this scheme, in an easy way, in existing computational codes that employ direct integration methods. Examples of dynamic solid bodies are presented, ilustrating the technique and library usage in existing applications. / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Este trabalho introduz uma estratégia de adaptação no tempo que utiliza um estimador de refinamento baseado no indicador geométrico curvatura. A metodologia desenvolvida é aplicada ao problema de integração numérica temporal, presente, por exemplo, no estudo do comportamento de corpos submetidos a cargas dinâmicas. O estimador de refinamento formulado demanda pouco esforço computacional, sendo facilmente aplicado aos diversos métodos de integração direta existentes. Visando a interação com esses métodos, constrói-se uma biblioteca orientada a objetos que incorpora a técnica de adaptação desenvolvida. A idéia principal desta ferramenta é prover, de forma fácil, a técnica de adaptação concebida a códigos computacionais existentes e que fazem uso dos referidos métodos de integração. Apresentam-se exemplos de dinâmica de corpos sólidos que ilustram o potencial de utilização da técnica e o uso da biblioteca em aplicações existentes.

We study a two-scale coupled system consisting of a macroscopic elliptic equation and a microscopic parabolic equation. This system models the interplay between a gas and liquid close to equilibrium within a porous medium with distributed microstructures. We use formal homogenization arguments to derive the target system. We start by proving well-posedness and inverse estimates for the two-scale system. We follow up by proposing a Galerkin scheme which is continuous in time and discrete in space, for which we obtain well-posedness, a priori error estimates and convergence rates. Finally, we propose a numerical error reduction strategy by refining the grid based on residual error estimators.

Made available in DSpace on 2016-06-02T20:05:58Z (GMT). No. of bitstreams: 1 DissFEAM.pdf: 568946 bytes, checksum: 953e676ad364fe570f94bc6a67fa2f34 (MD5) Previous issue date: 2006-05-31 / In this work we revise the main basic concepts on survival analysis and reliability. We present the most known nonparametric estimators of the survival function in the presence of censured data. Their estimates are calculed and compared in an real data set on human renal transplants. We point some problems that are not usually considered in the structure of these estimators. As consequense of such problems we can observe possible distortions in the estimates the direction of sub or over estimating the main characteristics of interest. Some alternatives are pointed out for finding lesse vulnerable estimators. Moreover, we propose a new modified piece wise exponential estimator, pointing out its properties. / Neste trabalho, revisamos os principais conceitos básicos de análise de sobre- vivência e confiabilidade, sob a abordagem da inferência clássica. Relacionamos os mais conhecidos estimadores não paramétricos das funções de sobrevivência, com dados cen- surados. Suas estimativas são calculadas e comparadas em um exemplo com dados reais de transplantes renais humanos. Apontamos alguns problemas que, muitas vezes, deixaram de ser considerados na estruturação destes estimadores. Como conseqüência destes pro- blemas, citamos possíveis distorções nas estimativas, no sentido de sub ou super estimar as principais características de interesse, e levantamos as possibilidades de se estar: expondo a riscos desnecessários pacientes, acompanhados em estudos clínicos, ou submetendo-se empresas a prejuízos financeiros, em estudos de fidelidade de clientes. Além disso, propomos um novo estimador modificado do tipo exponencial por partes, pela mudança que realizamos na taxa de risco acumulada de Kitchin nos subin- tervalos formados pelos tempos consecutivos das ocorrências dos eventos de interesse. Finalmente, dentro do conhecimento atualmente disponível, apontamos alternativas na busca de correções para alguns destes problemas e realçamos a necessidade da elaboração de estimadores menos vulneráveis a eles.

Uma questão interessante na política monetária é se os Bancos Centrais dão pesos iguais para desvios positivos e negativos da inflação e do hiato do produto das suas respectivas metas. Para responder à esta questão, estimou-se a função perda da autoridade monetária não parametricamente através do método de sieve estimator, expandindo-a através de uma base composta de polinômios ortogonais. A economia foi modelada com agentes foward-looking e com comprometimento por parte da autoridade monetária. O método foi aplicado para a os Estados Unidos desde 1960 e para o Brasil a partir de 1999. Para a economia norte-americana não foram encontradas evidências de assimetria nas preferências da autoridade monetária. Já no Brasil o Banco Central mostrou ter preferências assimétricas quanto à inflação, auferindo uma maior perda de desvios negativos do que positivos em relação à meta. / An interesting question in monetary policy is whether the Central Bank gives equal weights to positive and negative deviations of inflation and output gap from their targets. Trying answering this question, we estimated the monetary authority’s loss function nonparametrically, using the method of sieves, expanding it with orthogonal polynomials. The economy was model with forward-looking agents and with commitment of the monetary authority. We applied the method to U.S. monetary policy since 1960 and for Brazil since 1999. For the U.S. economy, it was not found evidence of asymmetry in the preferences of the monetary authority. In Brazil, the Central Bank proved to have asymmetric preferences about inflation, with a greater loss for negative deviations of inflation from the target rather for positive ones.