Thesis (Ph.D.)--Ohio University, June, 2003. / Title from PDF t.p. Includes bibliographical references (leaves 177-191)

Orientador: Hildete Prisco Pinheiro / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Matemática, Estatística e Computação Científica / Made available in DSpace on 2018-08-24T02:10:38Z (GMT). No. of bitstreams: 1 Argenton_JulianaLuzPassos_M.pdf: 2087574 bytes, checksum: b6014c3478501f5128fd13ddf952e6dd (MD5) Previous issue date: 2013 / Resumo: O objetivo deste trabalho é apresentar uma metodologia de árvore de regressão para dados censurados e correlacionados. O conjunto de dados analisado foi obtido a partir de uma pesquisa realizada entre Dezembro de 2005 e Janeiro de 2006, que entrevistou 119 famílias (1712 indivíduos) que vivem no pequeno vilarejo de Baependi, no Estado de Minas Gerais. São apresentadas duas metodologias com base no modelo de riscos proporcionais, a primeira desconsidera a possível correlação existente entre os indivíduos de uma mesma família e usa a primeira iteração da estimativa da verossimilhança completa nas divisões dos nós. Na segunda metodologia apresentada, a correlação entre os indivíduos de uma mesma família é incorporada no modelo de riscos proporcionais através de uma variável de fragilidade com distribuição Gama, neste caso o valor da estatística Escore é usado para escolher a melhor divisão dos nós. O objetivo da análise é avaliar as variáveis que aumentam o risco de apresentar hipertensão, diabetes tipo II e colesterol alto, que são os três principais fatores que aumentam o risco de doenças no coração. As variáveis respostas são as idades de diagnóstico desses fatores de risco. A censura é definida de acordo com a observação da idade do indivíduo no momento do diagnóstico da doença e a idade do indivíduo no momento da pesquisa. Desta forma, uma idade de diagnóstico maior que a idade no momento da pesquisa caracteriza a censura. / Abstract: The objective of this work is to present methods of regression trees for censored and correlated data. The dataset analyzed was obtained from a survey, in which 119 families (1712 individuals) living in Baependi village, in the Brazilian state of Minas Gerais, were interviewed. Two methodologies based on the proportional hazard model are presented. The first disregards the possible correlation among the individuals of the same family, using the first step of a full likelihood estimation procedure for splitting nodes. In the second methodology, the correlation among the individuals of the same family is incorporated in the proportional hazard model through a frailty variable with Gamma distribution. In this case, the value of the Score statistic is used for choosing the best splitting node. The main purpose of the analysis is to evaluate the variables that increase the risk of hypertension, type II diabetes and high cholesterol, which are the top three main factors that increase the risk of heart conditions. The response variables are the age-of-onset of these risk factors. Censoring is defined by observing the individual's age-of-onset at the moment of diagnosis and also at the moment of the survey. This way, an age-of-onset higher than the age at the moment of the survey indicates censoring. / Mestrado / Estatistica / Mestra em Estatística

Orientador: Renato da Rocha Lopes / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Elétrica e de Computação / Made available in DSpace on 2018-08-21T12:31:48Z (GMT). No. of bitstreams: 1 Barros_TiagoTavaresLeite_M.pdf: 6650333 bytes, checksum: 1abff9ddc2ee8bb4334ec96ba469f422 (MD5) Previous issue date: 2012 / Resumo: Nesta dissertação, nós discutimos o cálculo de funções de coerência de alta resolução para a estimação dos parâmetros de empilhamento em processamento de sinais sísmicos. Nosso foco é o algoritmo de estimação por Classificação de Sinais Múltiplos (MUSIC, do inglês MUltiple SIgnal Classification). Este pode superar a tradicional função de coerência semblance em casos em que há reflexões próximas ou interferentes. Nossa principal contribuição é a proposta de diversas simplificações para sua implementação. Primeiro, mostramos como obter os valores da função MUSIC a partir do subespaço de sinais do dado sísmico, que possui dimensão menor do que o subespaço de ruído, usualmente empregado. Depois disso, mostramos como obter o subespaço de sinais a partir do método da potência. Chamamos esta técnica de MUSIC com Método da Potência (PM-MUSIC). Também propusemos uma nova maneira de obtenção do espectro de MUSIC, baseada na decomposição em autovalores e autovetores da matriz de correlação temporal do dado sísmico. Este método contrasta com os algoritmos presentes na literatura, que se baseiam na correlação espacial. A partir do uso do Método da Potência, obtivemos reduções de complexidade tanto para a variante espacial quanto para a temporal do algoritmo MUSIC. Finalmente, também propusemos uma nova função de normalização para o cálculo de MUSIC, a qual chamamos de ponderação por semblance. Esta função leva em conta o espectro de velocidades obtido com a função de coerência semblance e lida com a alta variação dinâmica produzida pelo espectro de velocidades calculado com MUSIC. Nós comparamos a implementação de PM-MUSIC, a partir das correlações temporal e espacial. Exemplos numéricos com dados sísmicos sintéticos e de levantamentos reais demonstraram que o algoritmo PM-MUSIC supera o semblance e que sua variante temporal possui alta resolução, assim como sua variante espacial. Além disso, PM-MUSIC obtido a partir da correlação temporal mostrou-se extremamente robusto ao lidar com sinais correlacionados / Abstract: In this dissertation we discuss high-resolution coherence functions for the estimation of the stacking parameters in seismic signal processing. We focus on the MUltiple SIgnal Classification (MUSIC) algorithm, which uses the eigendecomposition of the seismic data to measure the coherence. MUSIC can outperform the traditional semblance in cases of close or interfering reflections. Our main contribution is to propose several simplifications to the implementation of MUSIC. First, we show how to compute MUSIC coherence measure in terms of the signal subspace of seismic data, which has lower dimension than the one currently used, the noise subspace. After that, we show how to obtain the signal subspace, iteratively, with the power method. We called this technique of Power Method MUSIC (PM-MUSIC). We also propose a new way to obtain the MUSIC spectrum, based on the eigendecomposition of the temporal correlation matrix of the seismic data. This is in contrast to the algorithms in the literature, which are based on the spatial correlation. Complexity reductions are obtained and discussed with the use of Power Method for both spatial and temporal variant of MUSIC. Finally, we propose a new normalization function for MUSIC, which we called semblance weighting. This function takes into account semblance coefficient and deals with high dynamic range in MUSIC velocity spectrum. We compared spatial and temporal correlation matrices, implemented with PM-MUSIC. Numerical examples with synthetic and real seismic data indicated that PM-MUSIC outperforms semblance and that the temporal variant of PM-MUSIC can present the same high-resolution as its spatial counterpart. Moreover, temporal PM-MUSIC is particularly useful when dealing with correlated signals / Mestrado / Telecomunicações e Telemática / Mestre em Engenharia Elétrica

This study dealt with shrinkage in multiple correlation coefficients computed for sample data when these coefficients are compared to the multiple correlation coefficients for populations and the effect of the ratio of utilized predictors to original predictors on the shrinkage in R square. The study sought to provide the rationale for selection of the shrinkage formula when the correlations between the predictors and the criterion are known and determine which of the three shrinkage formulas (Browne, Darlington, or Wherry) will yield the R square from sample data that is closest to the R square for the population data.

From Introduction: In geography during the 1950's there was a definite move away from the study of unique phenomena to the study of generalized phenomena or pattern (Mather and Openshaw, 1974). At the same time interrelationships between phenomena distributed in space and time became the topic of much interest among geographers, as well as members of other disciplines. The changing emphasis initiated acceptance of certain scientific principles (Cole, 1973), and mathematical techniques became the recognized and respected means through which objective analysis of pattern, structure, and interrelationships between a really distributed phenomena could be achieved (Ackerman, 1972; Burton, 1972; Gould, 1973). Geographers, as do members of other disciplines, frequently borrow mathematical techniques developed for problems encountered in the pure sciences and apply these techniques to what are felt to be analogous situations in geography.

This dissertation: (a) investigated the degree to which the squared canonical correlation coefficient is biased in multivariate nonnormal distributions and (b) identified formulae that adjust the squared canonical correlation coefficient (Rc2) such that it most closely approximates the true population effect under normal and nonnormal data conditions. Five conditions were manipulated in a fully-crossed design to determine the degree of bias associated with Rc2: distribution shape, variable sets, sample size to variable ratios, and within- and between-set correlations. Very few of the condition combinations produced acceptable amounts of bias in Rc2, but those that did were all found with first function results. The sample size to variable ratio (n:v)was determined to have the greatest impact on the bias associated with the Rc2 for the first, second, and third functions. The variable set condition also affected the accuracy of Rc2, but for the second and third functions only. The kurtosis levels of the marginal distributions (b2), and the between- and within-set correlations demonstrated little or no impact on the bias associated with Rc2. Therefore, it is recommended that researchers use n:v ratios of at least 10:1 in canonical analyses, although greater n:v ratios have the potential to produce even less bias. Furthermore,because it was determined that b2 did not impact the accuracy of Rc2, one can be somewhat confident that, with marginal distributions possessing homogenous kurtosis levels ranging anywhere from -1 to 8, Rc2 will likely be as accurate as that resulting from a normal distribution. Because the majority of Rc2 estimates were extremely biased, it is recommended that all Rc2 effects, regardless of which function from which they result, be adjusted using an appropriate adjustment formula. If no rationale exists for the use of another formula, the Rozeboom-2 would likely be a safe choice given that it produced the greatest number of unbiased Rc2 estimates for the greatest number of condition combinations in this study.

Thesis (Ph.D.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering, 1981. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND ENGINEERING. / Bibliography: leaves 74-76. / by Stephen W. Lang. / Ph.D.

This thesis provides statistical analysis methods and a validation procedure for conducting this statistical analysis, under the common random number (CRN) correlation-induction strategy. The proposed statistical analysis provides estimates for the unknown parameters that are needed for validating the model. While conducting this statistical analysis, we make some key assumptions. Validation comprises of a three-stage statistical procedure. The first stage tests for the multivariate normality,the second stage tests the structure of the covariance matrix between responses, and the third stage tests for the adequacy of the proposed model. The statistical analysis and validation procedures are illustrated with an example of a hospital simulation study. / Master of Science

This thesis investigates statistical correlation as a means to enhance the use of ground-based radar in analyzing satellite-path millimeter wave propagation through rain and ice crystals. The technique presented involves correlating dB values of the satellite signal attenuation and polarization isolation with dBZ values of radar backscatter from each of 128 range gates. In it, Pearson product moment correlation coefficients are calculated for attenuation and backscatter and for isolation and backscatter. When these coefficients are plotted versus radar range, one of four certain characteristic patterns usually appears. Ice-crystal depolarization produces a pattern of near zero attenuation coefficients and varying isolation coefficients. Rain produces a correlation pattern in which the attenuation coefficient pattern is nearly mirror image of the isolation coefficient pattern. A special case of the rain event occurs when the cross-polarized satellite signal is essentially constant. The correlation patterns for this case are exact mirror images. Rain attenuation and depolarization accompanied by additional depolarization from another source, produce a correlation pattern that is not symmetrical. This is due to the additional depolarization. Discussion of these expected patterns and examples of each are presented. / Master of Science

Thesis (M.S.)--State University of New York at Binghamton, Thomas J. Watson School of Engineering and Applied Science, Department of Computer Science, 2009. / Includes bibliographical references.