Orientadores: Caio Lucidius Naberezny Azevedo, Filidor Edilfonso Vilca Labra / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Matemática Estatística e Computação Científica / Made available in DSpace on 2018-08-26T22:33:37Z (GMT). No. of bitstreams: 1 Chaves_NathaliaLima_M.pdf: 3044792 bytes, checksum: 8fea3cd9d074997b605026a7a4526c35 (MD5) Previous issue date: 2015 / Resumo: A classe de modelos Birnbaum-Saunders (BS) foi desenvolvida a partir de problemas que surgiram na área de confiabilidade de materiais. Tais problemas, em geral, são ligados ao estudo de fadiga de materiais. No entanto, nos últimos tempos, essa classe de modelos tem sido aplicada em áreas fora do referido contexto como, por exemplo, em ciências da saúde, ambiental, florestal, demográficas, atuariais, financeira, entre outras, devido à sua grande versatilidade. Neste trabalho desenvolvemos a distribuição Birnbaum-Saunders (BS) baseada na normal assimétrica padrão sob a parametrização centrada (BSNAC) que, além de representar uma extensão da distribuição BS usual, apresenta diversas vantagens em relação à distribuição BS baseada na distribuição normal assimétrica sob a parametrização usual. Desenvolvemos também um modelo de regressão linear log-Birnbaum-Saunders. Apresentamos, tanto para a distribuição BSNAC quanto para o respectivo modelo de regressão, diversas propriedades. Desenvolvemos procedimentos de estimação sob os enfoques frenquentista e bayesiano, bem como ferramentas de diagnóstico para os modelos propostos, contemplando análise residual e medidas de influência. Realizamos estudos de simulação, considerando diferentes cenários, com o intuito de comparar as estimativas frequentistas e bayesianas, bem como avaliar o desempenho das medidas de diagnóstico. A metodologia aqui proposta foi ilustrada tanto com dados provenientes de estudos de simulação, quanto com conjuntos de dados reais / Abstract: The class of Birnbaum-Saunders (BS) models was developed from problems that arose in the field of material reliability. These problems generally are related to the study of material fatigue. However, in the last years, this class of models has been applied in areas outside that context, such as in health sciences, environmental, forestry, demographic, actuarial, financial, among others, due to its great versatility. In this work, we developed the skew-normal Birnbaum-Saunders distribution under the centered parameterization (BSNAC), which also represents an extension of the usual BS distribution and presents several advantages over the BS distribution based on the skew-normal distribution under the usual parameterization. We also developed a log-Birnbaum-Saunders linear regression model. We present several properties of both BSNAC distribution and the related regression model. We develop estimation procedures under the frequentist and Bayesian approaches, as well as diagnostic tools for the proposed models, contemplating residual analysis and measures of influence. We conducted simulation studies considering different scenarios, in order to compare the frequentist and Bayesian estimates and evaluate the performance of diagnostic measures. The methodology proposed here is illustrated with data sets from both simulation studies and real data sets / Mestrado / Estatistica / Mestra em Estatística

PhD (Statistics) / Department of Statistics / There has been a growing interest in investment opportunities in Africa. The net foreign direct investment (FDI) to Sub-Saharan Africa has increased from $13 billion in 2004 to about $54 billion in 2015. Investing on the stock markets is one of such investment opportunities. Stock markets in Africa have realised growth in market capitalization, membership, value and volume traded due to an increase in investments. This level of growth in African stock markets has raised questions about their efficiency. This thesis examined the weak-form informational efficiency of African stock markets. The aim therefore of this thesis is to test the efficiency of African stock markets in the weak-form of the Efficient Market Hypothesis (EMH) for eight countries, namely, Botswana, Egypt, Kenya, Mauritius, Morocco, Nigeria, South Africa and Tunisia. Since, the researcher will be testing the weak-form of the EMH, the data to be used is on past price information on the markets of the eight countries. Data for the eight countries were obtained from DataStream for the period between August 28, 2000 to August 28, 2015. The data is for a period of 180 months which resulted in 3915 data points. Although there have been studies on the weak-form market efficiency of African stock markets, the efficiency conclusions on the markets have been mixed. This problem might be due to the methods used in the analyses. First, most of the methods used were linear in nature although the data generating process of stock market data is nonlinear and hence nonlinear methods maybe more appropriate in its analysis. Also these linear methods tested the efficiency of African markets in absolute form, however, an efficiency conclusion relying solely on absolute efficiency might be misleading because, stock markets become efficient with time due to improvements in the quality of information processing from reforms on the markets. The researcher solved this problem of using absolute frequency by comparing the results when the presence of long-memory in frequency and time domains of the markets were examined. The researcher used a semi-parametric estimator, the Local Whittle estimator to test for long-memory in frequency domain and the Detrended Fluctuation Analysis (DFA) to test for long-memory in time domain. The DFA method is suitable for both stationary and nonstationary time series which makes it to have more power over methods like the rescaled range analysis (R/S) in the estimation of Hurst exponent. Second, the researcher examined whether the markets were predictable under the Adaptive Market Hypothesis (AMH). The researcher employed the Generalised Spectral (GS) test to examine the Martingale difference hypothesis (MDH) of the markets. The Generalised spectral (GS) test is a non-parametric ii test designed to detect the presence of linear and nonlinear dependencies in a stationary time series. The GS test considers dependence at all lags. Third, because of the nonlinear nature in the data-generating process on the markets, the stationarity of the market returns under a nonlinear Exponential Smooth Threshold Autoregressive (ESTAR) model was examined. A nonlinear ADF unit root test against ESTAR and a modified Wald-type test against ESTAR in the analysis were employed. Fourth, the self-exciting threshold Autoregressive (SETAR) method was employed to model the returns when non-linear patterns were observed as a result of nonlinear data generating process on the markets. The literature on market efficiency of African stock markets has shown that variations exist in the study characteristics. There are variations in the method of analysis, type of test, type of data employed, time period chosen and the scope of analysis for the studies. The researcher therefore quantitatively reviewed previous studies by means of meta-analysis to identify which study characteristics affects efficiency conclusions of African markets using the mixed effects model. The findings showed the presence of long-memory in the returns of the stock markets when the whole sample was used. This made the markets weak-form inefficient, however, when the researcher tested for the persistence of long-memory through time, there were periods the markets were efficient in the weak-form. The memory effect was low in the South African market but high in the Mauritian market. Furthermore, it was observed that, the returns for Egypt, which were highly predictable when the whole data was analysed became not highly predictable when the rolling window approach of the GS test was used. Egypt had one of the lowest percentages of the windows that had a p-value less than 0.05 after South Africa. The results obtained from using the non-linear unit root tests on the logarithmic price series of the markets under study showed that, the markets were non-stationary and hence weak-form efficient under an ESTAR framework but for Botswana. Thus the markets were weak-form efficient when analysed using a non-linear method. This observation means that Africa’s foreign direct investment would have been increased over the years if the appropriate methods are used. This is because, over the years, studies on the weak-form efficiency African stock markets have ended with mixed conclusions with most of the markets being concluded to be weak-form inefficient as a result of the use of linear methods in the analysis. This finding, to us, has had an effect on investors commitments to Africa because the right methodology was not employed. iii The findings from modelling the returns under the non-linear SETAR model showed that, the SETAR model performs better than the standard AR(1) and AR(2) model for all the markets under study after the non-linear patterns were identified in the returns series. The SETAR (2,2,2) model is a threshold model, therefore, investors are able to move freely in search of higher opportunities between the low and high regimes. Investors main aim is to make profits, hence, the threshold model of SETAR gives them the freedom to move to a regime where the rate of returns is increasing unlike the standard AR(1) and AR(2) linear models where there are no switching of regimes. Finally, none of the study characteristics in the market efficiency studies was found to be significant in efficiency conclusions of African stock markets but the indicator for publication bias was significant. This means that there has been a change in attitude in recent years towards studies on informational market efficiency whose results do not support the Efficient Market Hypothesis (EMH), unlike the earlier years when the EMH was formulated and acclaimed to be one of the best propositions in economics. It was therefore concluded that when time-varying methods are used in analysing weak-form efficiency, the dynamics of the markets become known to investors for proper decision-making. Also, nonlinear methods should be used in order to reflect the nonlinear nature of data capturing on the stock markets / NRF

Le rayonnement d'un haut-parleur monté sur une enceinte est généralement caractérisé par des mesures ou une simulation par la méthode des éléments finis. Cependant, ces méthodes de référence restent très coûteuses et ne permettent pas une interprétation physique des résultats. Dans ce manuscrit, nous proposons deux modèles analytiques pour prédire ce rayonnement, dans le cadre d'une application à une barre de son. Le premier modèle consiste à assimiler la géométrie de la barre de son à un sphéroïde. De cette manière, les variables sont séparées et on peut trouver une solution analytique sous la forme d'une somme d’harmoniques sphéroïdales. On décrira chaque étape de la méthode, ainsi que la comparaison des résultats à ceux des méthodes de référence. On développera particulièrement le calcul pour un haut-parleur circulaire sur le sphéroïde et la mise en place d'un critère de troncature des harmoniques. Ce modèle fonctionne bien en basse-fréquence, mais ne peut pas rendre compte de tous les phénomènes de diffraction par l'enceinte en haute fréquence. Pour celles-ci, la diffraction du champ sonore par les arêtes de l'enceinte devient non négligeable. Nous avons donc développé un second modèle analytique, basé sur une formulation intégrale de cette diffraction. Celle-ci est vue comme un ensemble de sources secondaires localisées sur les arêtes. On montrera comment établir ce modèle et on donnera des détails sur son implémentation. L'application de ce modèle permet d'interpréter physiquement le rayonnement d'une barre de son, et donc de conclure sur la validité du modèle sphéroïdal. On montre également comment les arêtes du baffle d'une enceinte entraînent des effets d'interférences constructives, qui peuvent induire un gain allant jusqu’à 3 dB. Ces effets entraînent également un phénomène contre-intuitif : si le haut-parleur n'est pas centré sur le baffle, la direction du maximum d'intensité du rayonnement tend à se décaler vers le côté opposé au décalage du haut-parleur sur le baffle. / The radiation from a loudspeaker mounted on a box shaped enclosure is usually characterized by measurements or simulations based on finite element method. However, these reference methods are still very expensive and do not allow for a physical interpretation of the results. In this manuscript, two analytical models are proposed to predict this radiation, within the framework of the application to a sound bar. The first model consists of assimilating the sound bar geometry to a spheroid. In this way, the variables are separated and an analytical solution can be found in the form of a sum of spheroidal harmonics. The calculation for a circular loudspeaker on the spheroid, as well as a harmonic truncation criterion, is detailed. Each step of the method will be described, as well as the comparison of the results with the reference methods. The calculation for a circular loudspeaker on the spheroid and the implementation of a harmonic truncation criterion will be developed. This model works well at low-frequency, but cannot account for the diffraction phenomena, which come from the enclosure and appear at high-frequency. Therefore, we have developed a second analytical model based on an integral formulation of the diffraction. This model is seen as a set of secondary sources located on the edges of the enclosure. It will be shown how to establish this model and how it can be implemented. This model allows us to interpret the radiation of a sound bar, and therefore allows us to conclude on the validity of the spheroidal model. It also shows how the baffle edges of a loudspeaker enclosure lead to constructive interference effects, resulting in a gain up to 3 dB. These effects also lead to a counterintuitive phenomenon: if the speaker is not centered on the baffle, the direction of maximum radiation intensity tends to shift to the opposite side of the speaker offset on the baffle.

Clinical drug trials are costly and time-consuming. Bayesian methods alleviate the inefficiencies in the testing process while providing user-friendly probabilistic inference and predictions from the sampled posterior distributions, saving resources, time, and money. We propose a dynamic linear model to estimate the mean response at each dose level, borrowing strength across dose levels. Our model permits nonmonotonicity of the dose-response relationship, facilitating precise modeling of a wider array of dose-response relationships (including the possibility of toxicity). In addition, we incorporate an adaptive approach to the design of the clinical trial, which allows for interim decisions and assignment to doses based on dose-response uncertainty and dose efficacy. The interim decisions we consider are stopping early for success and stopping early for futility, allowing for patient and time savings in the drug development process. These methods complement current clinical trial design research.

No description available.

No description available.

The Supreme Court of the United States has recently determined that assigning students to schools and classrooms based on racial identity is unconstitutional. However, it also left the door open for further and different rulings. If researchers are able to show that lack of consideration of race has deleterious effects on federally mandated programs and initiatives, the ruling may be modified or opened up to specific circumstances. Among its many consequences, this ruling brings a focus onto the question of student-teacher matching in classrooms. Over the years, there has been a great deal of discussion in the literature about matching teacher and student by race, ethnicity, gender, and language. Some people claim that matching is crucial for student success while others dispute this claim. The current study examines the question of racial and ethnic matching empirically in the context of a large-scale randomized controlled study of an innovation for middle school mathematics learners. It extends the literature by (1) focusing on the relationship between student-teacher match and a specific, heavily documented situation with targeted learning goals, (2) adding information about Hispanic students to the discussion, and (3) helping evaluate factors that may be important in determining the validity of large-scale experiments. Alone and in conjunction with other similar empirical evidence, it will also have a significant effect on federal and state educational policy. The sample consists of the 92 teachers and 1576 7th grade students on 76 school campuses throughout 8 Texas regions who participated in the Scaling-Up SimCalc project. Teachers and students either used SimCalc Mathworlds™ curriculum and technology or a control for a two-week replacement unit. The crux of the current analysis was a match between aggregated and individual teacher and student characteristics and an inquiry into how these matches influence student math performance in the classroom within and between our experimental and control group. Hierarchical Linear Modeling (HLM) analysis was used to investigate the differences in student mathematics performance, modeled as students nested in classrooms nested in schools. / Ph. D.

Prior research has shown that both lead users and opinion leaders may propel the diffusion of innovation. This raises the question of whether lead users and opinion leaders are positioned similarly in social networks, which we address using a sample of 23 school classes consisting of 537 children. Research among children is very scarce in this particular domain. Our statistical analyses based on hierarchical linear modeling reveal two general results: first, lead users among children appear to possess a variety of links between clusters; second, opinion leaders are locally positioned within clusters of children and have many direct links. (authors' abstract)

A identificação dos fatores que influenciam o processo de escolha do tomador na demanda por crédito apresenta não apenas um interesse mercadológico, mas também em termos acadêmicos e para o formulador de políticas públicas, buscando determinar os impactos de uma decisão que influencia o ambiente macroeconômico, bem como o comportamento dos agentes. Nestes termos, os modelos tradicionais de análise da demanda muitas vezes apresentam suposições pouco realistas e bastante restritivas, necessárias para o processo de estimação dos parâmetros de interesse. Os Modelos Lineares Generalizados Mistos com Variáveis Latentes (GLLAMM) constituem uma classe de modelos que abrangem os tradicionais modelos lineares generalizados e os modelos lineares generalizados mistos, possibilitando uma maior flexibilidade na combinação de um processo de escolha discreta com a determinação dos valores demandados de forma contínua, não impondo um processo único para todas as instituições analisadas, nem tão pouco do processo de escolha. Esta classe de modelos foi aplicada para estudar a demanda por empréstimos bancários utilizando-se informações de uma rica base de dados, a Central de Risco de Crédito do Banco Central. Assim, foi possível a identificação de variáveis como a duração da operação e a classificação de risco da operação que apresentam uma maior relevância no processo de escolha do banco, enquanto que outras, como as garantias, mostraram-se mais importante no volume a ser demandado. A identificação de nichos específicos de algumas instituições foi possível a partir desta análise. A flexibilidade desta classe de modelos também foi utilizada no intuito de se identificar os fatores que influenciam a demanda por crédito pelos bancos nas operações de redesconto, tendo conseguido tratar o problema de superdispersão ocasionado pelo excesso de zeros neste conjunto de dados. Adicionalmente, tanto efeitos diretos quanto indiretos da taxa de redesconto foram possíveis de serem estudados a partir da inclusão de efeitos aleatórios tanto no intercepto, possibilitando a incorporação de efeitos específicos de cada instituição financeira, bem como nos coeficientes, captando comportamentos individuais de cada banco frente a um mesmo estímulo. / The aim of this study is to identify the variables that affect borrower´s decision-making process through the estimation of loan demand equations. This research is relevant not only for market practitioners, but also to academics and to policy-makers, concerned with the evaluation of possible impacts of decision on the economic environment and on the agent´s behavior. Traditional models for demand estimation make unreasonable and very restrictive assumptions to estimate the parameters of interest. Generalized Linear Latent and Mixed Models (GLLAMM) constitute a class of models that includes the traditional Generalized Linear Models (GLM) and Generalized Linear Mixed Models (GLMM), which offer more flexibility and they are particularly suitable to situations that combine discrete choice with continuous decisions. Therefore, we can estimate individual equation for each bank simultaneously. This class of models was applied to study the demand for bank loans using a rich dataset provided by the Central Bank?s Credit Risk Bureau. Among the analyzed variables, the loan maturity and the credit risk classification were important in the bank choice while warranties were important in the decision related to the amount borrowed. Moreover, we could also identify specific market segments for some banks. The flexibility of this class of models was also used to identify the factors affecting the demand for overnight funds by commercial banks. This model overcomes overdispersion problems caused by excess of zeros found in the dataset. Additionally, we identified the direct and indirect effects of rediscount rate through the inclusion of random effects in the intercept (incorporating specific effects for each bank) and other coefficients (identifying individual behavior of each bank regarding the same stimuli).

Nesta dissertação apresentamos métodos de diagnóstico em modelos semiparamétricos sob erros normais, em especial os modelos semiparamétricos com uma variável explicativa não paramétrica, conhecidos como modelos lineares parciais. São utilizados splines cúbicos para o ajuste da variável resposta e são aplicadas funções de verossimilhança penalizadas para a obtenção dos estimadores de máxima verossimilhança com os respectivos erros padrão aproximados. São derivadas também as propriedades da matriz hat para esse tipo de modelo, com o objetivo de utilizá-la como ferramenta na análise de diagnóstico. Gráficos normais de probabilidade com envelope gerado também foram adaptados para avaliar a adequabilidade do modelo. Finalmente, são apresentados dois exemplos ilustrativos em que os ajustes são comparados com modelos lineares normais usuais, tanto no contexto do modelo aditivo normal simples como no contexto do modelo linear parcial. / In this master dissertation we present diagnostic methods in semiparametric models under normal errors, specially in semiparametric models with one nonparametric explanatory variable, also known as partial linear model. We use cubic splines for the nonparametric fitting, and penalized likelihood functions are applied for obtaining maximum likelihood estimators with their respective approximate standard errors. The properties of the hat matrix are also derived for this kind of model, aiming to use it as a tool for diagnostic analysis. Normal probability plots with simulated envelope graphs were also adapted to evaluate the model suitability. Finally, two illustrative examples are presented, in which the fits are compared with usual normal linear models, such as simple normal additive and partially linear models.