Doctor of Philosophy / Department of Statistics / Gary L. Gadbury / Multiple testing research has undergone renewed focus in recent years as advances in high throughput technologies have produced data on unprecedented scales. Much of the focus has been on false discovery rates (FDR) and related quantities that are estimated (or controlled for) in large scale multiple testing situations. Recent papers by Efron have directly addressed this issue and incorporated measures to account for high-dimensional correlation structure when estimating false discovery rates and when estimating a density. Other authors also have proposed methods to control or estimate FDR under dependencies with certain assumptions. However, not much focus is given to the stability of the results obtained under dependencies in the literature. This work begins by demonstrating the effect of dependence structure on the variance of the number of discoveries and the false discovery proportion (FDP). A variance of the number of discoveries is shown and the density of a test statistic, conditioned on the status (reject or failure to reject) of a different correlated test, is derived. A closed form solution to the correlation between test statistics is also derived. This correlation is a combination of correlations and variances of the data within groups being compared. It is shown that these correlations among the test statistics affect the conditional density and alters the threshold for significance of a correlated test, causing instability in the results. The concept of performing tests within networks, Conditional Network Testing (CNT) is introduced. This method is based on the conditional density mentioned above and uses the correlation between test statistics to construct networks. A method to simulate realistic data with preserved dependence structures is also presented. CNT is evaluated using simple simulations and the proposed simulation method. In addition, existing methods that controls false discovery rates are used on t-tests and CNT for comparing performance. It was shown that the false discovery proportion and type I error proportions are smaller when using CNT versus using t-tests and, in general, results are more stable when applied to CNT. Finally, applications and steps to further improve CNT are discussed.

Master of Science / Department of Statistics / Gary Gadbury / High-dimensional data (HDD) have been encountered in many fields and are characterized by a “large p, small n” paradigm that arises in genomic, lipidomic, and proteomic studies. This report used a simulation study that employed basic block diagonal covariance matrices to generate correlated HDD. Quantities of interests in such data are, among others, the number of ‘significant’ discoveries. This number can be highly variable when data are correlated. This project compared randomization tests versus usual t-tests for testing of significant effects across two treatment conditions. Of interest was whether the variance of the number of discoveries is better controlled in a randomization setting versus a t-test. The results showed that the randomization tests produced results similar to that of t-tests.

In this paper a resilient sensor network is built in order to lessen the effects of a small portion of corrupted sensors when an aggregated result such as the average needs to be obtained. By examining the variance in sensor readings, a change in the pattern can be spotted and minimized in order to maintain a stable aggregated reading. Offset in sensors readings are also analyzed and compensated to help reduce a bias change in average. These two analytical techniques are later combined in Kalman filter to produce a smooth and resilient average given by the readings of individual sensors. In addition, principal components analysis is used to detect variations in the sensor network. Experiments are held using real sensors called MICAz, which are use to gather light measurements in a small area and display the light average generated in that area.

This thesis analyzes the interdependence of stock markets in the Czech Republic, Hungary, Germany and the U.S. between 1994 and 2010. Its aim is to prove or disprove the hypothesis that the correlation of stock markets is growing. Furthermore, it aims to determine whether the benefits of international diversification are important in the 21st century. The first part defines the basic terms related to this issue. The next section describes the historical development of stock markets and major stock indexes. The following is the longest chapter, which deals with the dependence of equity markets. The degree of dependence between markets is measured by correlation coefficients. In order to calculate the correlation coefficients 3 methods were used: 12-month, 24-month moving window and the exponential weighting. It was confirmed that correlation increases in periods of high market volatility and if the equity markets go up. Correlation decreases in periods of low market volatility and if the stock markets go down. There is a long-term increasing trend in the correlation development. The last chapter describes the recent study of P. Christoffersen, V. Errunza, K. Jacobs, and X. Jin, which also deals with this issue. Although the authors used more complex mathematical and statistical methods, their conclusions are similar. At the end of this chapter there is a study dealing with the benefits of international diversification in 2005 and 2009. Its conclusion is that the benefits of international diversification are disappearing.

Thesis advisor: Zhijie Xiao / In recent years, quantile regression has achieved increasing prominence as a quantitative method of choice in applied econometric research. The methodology focuses on how the quantile of the dependent variable is influenced by the regressors, thus providing the researcher with much information about variations in the relationship between the covariates. In this dissertation, I consider two quantile regression models where the information set may contain quantiles of the regressors. Such frameworks thus capture the dependence between quantiles - the quantile of the dependent variable and the quantile of the regressors - which I call models of quantile dependence. These models are very useful from the applied researcher's perspective as they are able to further uncover complex dependence behavior and can be easily implemented using statistical packages meant for standard quantile regressions. The first chapter considers an application of the quantile dependence model in empirical finance. One of the most important parameter of interest in risk management is the correlation coefficient between stock returns. Knowing how correlation behaves is especially important in bear markets as correlations become unstable and increase quickly so that the benefits of diversification are diminished especially when they are needed most. In this chapter, I argue that it remains a challenge to estimate variations in correlations. In the literature, either a regime-switching model is used, which can only estimate correlation in a finite number of states, or a model based on extreme-value theory is used, which can only estimate correlation between the tails of the returns series. Interpreting the quantile of the stock return as having information about the state of the financial market, this chapter proposes to model the correlation between quantiles of stock returns. For instance, the correlation between the 10th percentiles of stock returns, say the U.S. and the U.K. returns, reflects correlation when the U.S. and U.K. are in the bearish state. One can also model the correlation between the 60th percentile of one series and the 40th percentile of another, which is not possible using existing tools in the literature. For this purpose, I propose a nonlinear quantile regression where the regressor is a conditional quantile itself, so that the left-hand-side variable is a quantile of one stock return and the regressor is a quantile of the other return. The conditional quantile regressor is an unknown object, hence feasible estimation entails replacing it with the fitted counterpart, which then gives rise to problems in inference. In particular, inference in the presence of generated quantile regressors will be invalid when conventional standard errors are used. However, validity is restored when a correction term is introduced into the regression model. In the empirical section, I investigate the dependence between the quantile of U.S. MSCI returns and the quantile of MSCI returns to eight other countries including Canada and major equity markets in Europe and Asia. Using regression models based on the Gaussian and Student-t copula, I construct correlation surfaces that reflect how the correlations between quantiles of these market returns behave. Generally, the correlations tend to rise gradually when the markets are increasingly bearish, as reflected by the fact that the returns are jointly declining. In addition, correlations tend to rise when markets are increasingly bullish, although the magnitude is smaller than the increase associated with bear markets. The second chapter considers an application of the quantile dependence model in empirical macroeconomics examining the money-output relationship. One area in this line of research focuses on the asymmetric effects of monetary policy on output growth. In particular, letting the negative residuals estimated from a money equation represent contractionary monetary policy shocks and the positive residuals represent expansionary shocks, it has been widely established that output growth declines more following a contractionary shock than it increases following an expansionary shock of the same magnitude. However, correctly identifying episodes of contraction and expansion in this manner presupposes that the true monetary innovation has a zero population mean, which is not verifiable. Therefore, I propose interpreting the quantiles of the monetary shocks as having information about the monetary policy stance. For instance, the 10th percentile shock reflects a restrictive stance relative to the 90th percentile shock, and the ranking of shocks is preserved regardless of shifts in the shock's distribution. This idea motivates modeling output growth as a function of the quantiles of monetary shocks. In addition, I consider modeling the quantile of output growth, which will enable policymakers to ascertain whether certain monetary policy objectives, as indexed by quantiles of monetary shocks, will be more effective in particular economic states, as indexed by quantiles of output growth. Therefore, this calls for a unified framework that models the relationship between the quantile of output growth and the quantile of monetary shocks. This framework employs a power series method to estimate quantile dependence. Monte Carlo experiments demonstrate that regressions based on cubic or quartic expansions are able to estimate the quantile dependence relationships well with reasonable bias properties and root-mean-squared errors. Hence, using the cubic and quartic regression models with M1 or M2 money supply growth as monetary instruments, I show that the right tail of the output growth distribution is generally more sensitive to M1 money supply shocks, while both tails of output growth distribution are more sensitive than the center is to M2 money supply shocks, implying that monetary policy is more effective in periods of very low and very high growth rates. In addition, when non-neutral, the influence of monetary policy on output growth is stronger when it is restrictive than expansive, which is consistent with previous findings on the asymmetric effects of monetary policy on output. / Thesis (PhD) — Boston College, 2009. / Submitted to: Boston College. Graduate School of Arts and Sciences. / Discipline: Economics.

O objetivo do presente trabalho é estudar a correlação entre as taxas de produção gasosa e as magnitudes heliocêntricas do cometa C/1995 O1 (Hale-Bopp), tanto na fase pré-periélica como na fase pós-periélica. As evoluções da magnitude e das taxas de produção gasosa de H2O (água), OH (radical hidroxila) e CO (monóxido de carbono), ao longo da aproximação e do afastamento do cometa em relação ao Sol, são analisadas. Para essa análise, foram utilizadas 11.734 estimativas de magnitudes visuais, extraídas do ICQ (International Comet Quarterly) e 88 observações do monóxido de carbono (Biver, comunicação particular (2007); Disanti et al. 2001; Jewitt et al. 1996), cobrindo o intervalo de distâncias heliocêntricas de rh = 7,464 UA (na fase pré-periélica) até rh = 14,070 UA (na fase pós-periélica). É mostrado que a atividade do Hale-Bopp (temperatura média superficial ~ 110 K), além de 6,3 UA do Sol é controlada pela emissão do CO (temperatura de sublimação ~ 24 K), antes que pela emissão da H2O (temperatura de sublimação ~ 152 K). Esse resultado é consistente com as observações em ondas milimétricas de Biver et al. 1996 e Jewitt et al. 1996, realizadas em 6,5 UA. / The purpose of the present work is to study the correlation between the gas production rates and heliocentric magnitudes of comet C/1995 O1 (Hale-Bopp), in the pre-perihelion phase as well as in the post-perihelion phase. The evolutions of magnitudes and gas production rates of H2O (water), OH (hydroxil radical) and CO (carbon monoxide), along the approach to and leave of the comet from the Sun, are analyzed. For this analysis, we used 11,734 visual magnitude estimates, extracted from ICQ (International Comet Quarterly) and 88 observations of carbon monoxide (Biver, private communication (2007); DiSanti et al. (2001); Jewitt et al. (1996)), covering the range of heliocentric distances from rh = 7.464 AU (in the pre-perihelion phase) to rh = 14.070 AU (in the post-perihelion phase). It is shown that the activity of Hale-Bopp (average surface temperature ~ 110 K) beyond 6.3 AU from the Sun is controlled by CO emission (sublimation temperature ~ 24 K) rather than by H2O (sublimation temperature ~ 152 K). This result is consistent with millimeter-wave observations of Biver et al. (1996) and Jewitt et al. (1996), made at 6.5 AU.

É bem conhecido que os intervalos proibidos de energia (gaps) de sólidos semicondutores e isolantes calculados resolvendo-se a equação de Kohn-Sham (KS) com alguma aproximação ao funcional de troca e correlação (XC) da teoria do funcional da densidade (density-functional theory, DFT) são geralmente muito pequenos quando comparados com os valores de gap obtidos experimentalmente. Estes erros, que podem atingir 100%, indicam uma falha nos funcionais de troca e correlação aproximados usados na DFT. Em particular, encontrou-se que o potencial XC, obtido através da derivada do funcional de energia XC, apresenta uma descontinuidade (xc ) quando um elétron extra é adicionado. Portanto, mesmo o gap KS exato não é idêntico ao gap verdadeiro, pois pode haver uma descontinuidade no funcional XC que tem de ser adicionado ao gap KS. Este trabalho propõe uma nova abordagem matemática para o cálculo da descontinuidade do funcional XC e a correção do gap. Inicialmente, as conseqüências desta nova abordagem foram estudadas para os 36 primeiros átomos da tabela periódica (do átomo de hidrogênio, H, ao átomo de criptônio, Kr), utilizando-se para isso os funcionais LDA, GGA e também funcionais híbridos, em combinação com diversos conjuntos de base. A partir da comparação entre valores da descontinuidade calculados teoricamente e resultados prévios da descontinuidade obtidos na literatura para os átomos de lítio (Li) e berílio (Be), foram escolhidos os melhores funcionais em combinação com as melhores bases. Num segundo passo, a nova metodologia para o cálculo da descontinuidade, com os melhores funcionais e bases eleitos, foi aplicada ao cálculo de fragmentos de poliacetileno. / It is well-known that the energy gaps of insulators and semiconductors calculated via the Kohn-Sham (KS) scheme with some approximation to the exchange-correlation (XC) functional of density-functional theory (DFT) are in general too small when compared to experimental gaps. These errors, which can be up to 100%, indicate a fault in the approximate exchange-correlation functionals currently available in DFT. Specifically, it was found that the XC potential, the functional derivative of the XC energy, shows a discontinuity (xc ) when an electron is added. Therefore, the exact KS gap is not identical to true gap, since there can be a discontinuity in the XC functional which should be added to KS gap. The present work proposes a new mathematical approach to the calculation of the discontinuity of the XC functional and the correction of the gap. In a first step, the consequences of this new approach were studied for the first 36 atoms of the periodic table (from Hydrogen, H, to Krypton, Kr), with the LDA and GGA functionals and also some hybrids, in combination with several kinds of basis set. From the comparison between calculated discontinuity values and previous results for the discontinuity reported in the literature for the Lithium (Li) and Beryllium (Be) atoms, the best functionals in combination with the best basis set were selected. In a second step, the new methodology with the best functionals/basis set was applied to the calculation of energy gaps of polyacetylene fragments.

Background: The policework is well known as physical demanding and often are controlled through a test-procedure. The Swedish police academy test-procedure has never been closely looked upon with a purpose to standardize its content. Purpose: This study aims to provide results from this test-procedure that can be used in a standardized and effective education in the Swedish police academy. The focus is on the number of trials given and the 14 tests used, combined with the development of physical capacity of the police students.Method: Descriptive data from the years 2007-08 and 2012-2016 has been analyzed with SPSS v25. A Pearson correlation, Spearman correlation, Wilcoxon signed rank test and pared t-test was conducted depending on the content of the data.Result: The performance of the second attempt seemed to be higher, especially in the Harres test. Alow correlation between the 14 tests was shown. The physical capacity of the students was increased through the education accept in the agility and aerob capacity.  Conclusion: The students increase their performance over the second attempts doe to a familiarization effect. It’s also showed that the test correlation is relatively low and that the students physical capacity increase during the police academy.

A avaliação da presença de compostos orgânicos voláteis nos efluentes líquidos é importante para atendimento à legislação vigente e proteção do sistema de esgotamento sanitário e da saúde ocupacional. Objetivos: o uso de técnica rápida de identificação e quantificação de compostos orgânicos voláteis do grupo BTEX presentes em amostras de efluentes líquidos através das emissões gasosas permite verificar quais delas não atendem à legislação vigente, através de faixas de concentração de interesse. A possibilidade de efetuar em campo essa avaliação promove maior agilidade nas ações de controle da qualidade do efluente bem como da exposição ocupacional. Também permite racionalizar os custos de monitoramento através da redução das amostras enviadas aos laboratórios. Materiais e métodos: os compostos avaliados causam agravos à saúde e, por serem altamente voláteis, possuem altos percentuais de distribuição no ar. O benzeno é comprovadamente cancerígeno, não existindo limites seguros para exposição a esse composto. A legislação brasileira fornece os subsídios necessários para o controle da produção, usos e disposição de BTEX. Há ações em andamento para um melhor controle e monitoramento da exposição ocupacional ao benzeno, em especial, e também ações para o monitoramento das populações expostas a ele. Com o objetivo de efetuar leituras no campo que identifiquem e quantifiquem os compostos do grupo BTEX, optou-se pela utilização de espectrofotômetro de infravermelho portátil. Resultados: os resultados das leituras das emissões desses compostos foram comparados com os resultados das análises de BTEX efetuadas nos efluentes líquidos. Para tanto, foram considerados os parâmetros existentes na legislação relativos à exposição ocupacional e à legislação de lançamento de efluentes. Conclusão: a avaliação dos resultados permite dizer que é factível a utilização da leitura das emissões gasosas para identificar a presença de BTEX em amostras de efluentes líquidos e que essas leituras podem quantificar valores que ultrapassem as faixas de concentração de interesse, através dos modelos de regressão identificados nessa avaliação. Para a construção de modelos de regressão mais confiáveis entre as concentrações das emissões gasosas e das concentrações no efluente líquido, são necessários estudos complementares. / The evaluation of the presence of volatile organic compounds in wastewater is important for compliance with current regulations, protection of the wastewater sewer systems and occupational health. Objectives: Identification and quantification of volatile organic compounds of BTEX group present in samples of liquid effluents through their gaseous emissions, allows checking which of them do not meet the current regulation, through concentration ranges of interest. The possibility of performing this evaluation in the field promotes greater flexibility in efforts to control the quality of the effluent as well as occupational exposure. It also allows streamlining the monitoring costs through reduction of liquid samples to be sent to laboratories. Materials and methods: The compounds evaluated cause health problems and for being highly volatile have high percentages of distribution in the air. Benzene is a proven carcinogen and there are no safe limits for exposure to this compound. Brazilian regulations provide the necessary subsidies for production control, uses and disposal of BTEX subsidies needed to control the production, uses and disposal of BTEX. There are ongoing activities to better control and monitoring of occupational exposure and also actions for monitoring populations exposed to benzene. An infrared spectrophotometer was used in air analyses to identify BTEX. Results: The results of the compounds emissions scans were compared with the results of BTEX analyzes in wastewater. For both parameters were considered in the regulation relating to occupational exposure and the rules of effluent discharge. Conclusion: The evaluation of the results allows saying that it is feasible to use gaseous emissions scans to identify the presence of BTEX in wastewater samples and they can indicate scans that exceed the concentration ranges of interest, through mathematical models this assessment. For mathematical models that provide precise curves of the correlation between the concentrations of gaseous emissions and liquid effluent concentrations, complementary studies are needed.

Embriologicamente, o tecido ósseo surge posteriormente aos músculos. A musculatura temporomandibular e facial apresentam-se definidas quando da ossificação da face e crânio, com os ossos se desenvolvendo dentro de um envelope de músculo e fascia. Quando contrações ativas se iniciam em um músculo, ele passa a exercer contínuo efeito sobre os ossos aos quais pertencem sua origem e inserção. O tecido ósseo é tido como um dos mais plásticos e responde prontamente às forças e tensões geradas pelos músculos. Evidenciase que os músculos, desde a fase pré-natal, desempenham papel na conformação dos ossos, e continuam a direcionar o crescimento pelo efeito de suas funções. Oito pacientes (4? e 4?) com idade entre 7 e 12 anos, com maloclusão dental classe III de Angle, foram submetidos a registros eletromiográficos do músculo masseter, subdividido em três regiões: superior, média e inferior, empregando-se o eletromiógrafo K6-I/EMG Eight Channel Surface Electromyograph (Myo-tronics Co., Seatle, WA, EUA), em 17 condições clínicas, e à análise cefalométrica através do programa Radiocef (RadioMemory, Floresta, BH, Brasil), de 15 medidas angulares e lineares. Os dados obtidos foram submetidos ao teste de correlação de Spearman, por meio do programa Statistica, release 4.3D (StatSoft, Inc., 1993, EUA). Correlações estatisticamente significantes foram encontradas entre as condições clínicas de repouso, propulsão livre da mandíbula, mastigação molar, oclusão cêntrica forçada, desvio lateral, deglutição de água, mastigação incisiva, fechamento contra-resistência e deglutição de saliva; e, entre as medidas cefalométricas, ângulo SNA, ângulo incisivo superior.NA, distância incisivo superior-NA, ângulo ANB, ângulo IMPA, distância Pog-NB, ângulo SN.Gn(Y), ângulo incisivo inferior.NB, ângulo interincisivo, distância incisivo inferior-NB, ângulo SN.GoGn, ângulo SN.Plano Oclusal, ângulo SNB e ângulo SND. As regiões superior, média e inferior do músculo masseter participaram em 32, 18 e 50% dos resultados estatisticamente significantes, respectivamente. Os resultados corroboram com hipóteses sobre a influência da musculatura no estabelecimento da maloclusão dental, e a relação entre a função dos músculos do sistema estomatognático e o desenvolvimento craniofacial precisa ser compreendida para se poder predizer o desenvolvimento craniofacial. / Embryologically, the bone tissue develops after muscle tissue. The temporomandibular and facial muscles are well defined when the cranial and facial ossifications occur, with bone structures developing inside of a muscle and fascia envelope. When active contractions begin in a muscle, it exerts an uninterrupted effect over bones to which belongs its origin and insertion. The bones tissues are one of the most plastic tissues, and interact readily to any tension and pressure developed by muscles. It\'s axiomatic that since previous to childbirth, muscles exert a role in bone structure development, and keep directing the skeletal growth with the effect of its functions. Eight patients (4? and 4?), between 7 and 12 years of age, with Angle\'s class III dental malocclusion were studied using electromyographic registers of the masseter muscle, divided in three regions: superior, middle and inferior, and cephalometric analysis. The K6-I/EMG Eight Channel Surface Electromyograph (Myo-tronics Co., Seatle, WA, EUA) was employed to get the registers in 17 clinical conditions, and the Radiocef (RadioMemory, Floresta, BH, Brasil) software was employed to get cephalometric analysis of 15 linear and angular measurements. The data obtained were submitted to statistical analysis by Spearman correlation test, with the Statistica, release 4.3D (StatSoft, Inc., 1993, EUA) software. Correlations were found among the clinical conditions of rest, free protraction of the jaw, molar chewing, centric occlusion, lateral movement without occlusal contact, swallowing water, incisal chewing, clenching the jaw against resistance and swallowing saliva; and among the following cephalometric measurements: the SNA angle, the Upper Incisor.NA angle, Upper Incisor-NA distance, ANB angle, IMPA angle, Pog-NB distance, SN.Gn(Y) angle, Lower Incisor.NB angle, Interincisors angle, Lower Incisor-NB distance, SN.GoGn angle, SN.Occlusal Plane angle, SNB angle and SND angle. The superior, middle and inferior regions of the masseter muscle were present in 32, 18 and 50% of the statistically significant results, respectively. The results reinforce the assumption of muscle influence over the development of dental malocclusion, and that the interaction of muscle function and skeletal growth must be understood to predict craniofacial development.