Keeping Variables within Bounds: Using Information between Observations

This research develops an econometric framework to analyze time series processes with bounds. The framework is general enough that it can incorporate several different kinds of bounding information that constrain continuous-time stochastic processes between discretely-sampled observations. It applies to situations in which the process is known to remain within an interval between observations, by way of either a known constraint or through the observation of extreme realizations of the process. The main statistical technique employs the theory of maximum likelihood estimation. This approach leads to the development of the asymptotic distribution theory for the estimation of the parameters in bounded diffusion models. The results of this analysis present several implications for empirical research. The advantages are realized in the form of efficiency gains, bias reduction and in the flexibility of model specification. A bias arises in the presence of bounding information that is ignored, while it is mitigated within this framework. An efficiency gain arises, in the sense that the statistical methods make use of conditioning information, as revealed by the bounds. Further, the specification of an econometric model can be uncoupled from the restriction to the bounds, leaving the researcher free to model the process near the bound in a way that avoids bias from misspecification. One byproduct of the improvements in model specification is that the more precise model estimation exposes other sources of misspecification. Some processes reveal themselves to be unlikely candidates for a given diffusion model, once the observations are analyzed in combination with the bounding information. A closer inspection of the theoretical foundation behind diffusion models leads to a more general specification of the model. This approach is used to produce a set of algorithms to make the model computationally feasible and more widely applicable. Finally, the modeling framework is applied to a series of interest rates, which, for several years, have been constrained by the lower bound of zero. The estimates from a series of diffusion models suggest a substantial difference in estimation results between models that ignore bounds and the framework that takes bounding information into consideration.

Statistical learning of random probability measures

The study of random probability measures is a lively research topic that has attracted interest from different fields in recent years. In this thesis, we consider random probability measures in the context of Bayesian nonparametrics, where the law of a random probability measure is used as prior distribution, and in the context of distributional data analysis, where the goal is to perform inference given avsample from the law of a random probability measure. The contributions contained in this thesis can be subdivided according to three different topics: (i) the use of almost surely discrete repulsive random measures (i.e., whose support points are well separated) for Bayesian model-based clustering, (ii) the proposal of new laws for collections of random probability measures for Bayesian density estimation of partially exchangeable data subdivided into different groups, and (iii) the study of principal component analysis and regression models for probability distributions seen as elements of the 2-Wasserstein space. Specifically, for point (i) above we propose an efficient Markov chain Monte Carlo algorithm for posterior inference, which sidesteps the need of split-merge reversible jump moves typically associated with poor performance, we propose a model for clustering high-dimensional data by introducing a novel class of anisotropic determinantal point processes, and study the distributional properties of the repulsive measures, shedding light on important theoretical results which enable more principled prior elicitation and more efficient posterior simulation algorithms. For point (ii) above, we consider several models suitable for clustering homogeneous populations, inducing spatial dependence across groups of data, extracting the characteristic traits common to all the data-groups, and propose a novel vector autoregressive model to study of growth curves of Singaporean kids. Finally, for point (iii), we propose a novel class of projected statistical methods for distributional data analysis for measures on the real line and on the unit-circle.

First stars XII. Abundances in extremely metal-poor turnoff stars, and comparison with the giants

Context. The detailed chemical abundances of extremely metal-poor (EMP) stars are key guides to understanding the early chemical evolution of the Galaxy. Most existing data, however, treat giant stars that may have experienced internal mixing later. Aims. We aim to compare the results for giants with new, accurate abundances for all observable elements in 18 EMP turno. stars. Methods. VLT/UVES spectra at R similar to 45 000 and S/N similar to 130 per pixel (lambda lambda 330-1000 nm) are analysed with OSMARCS model atmospheres and the TURBOSPECTRUM code to derive abundances for C, Mg, Si, Ca, Sc, Ti, Cr, Mn, Co, Ni, Zn, Sr, and Ba. Results. For Ca, Ni, Sr, and Ba, we find excellent consistency with our earlier sample of EMP giants, at all metallicities. However, our abundances of C, Sc, Ti, Cr, Mn and Co are similar to 0.2 dex larger than in giants of similar metallicity. Mg and Si abundances are similar to 0.2 dex lower (the giant [Mg/Fe] values are slightly revised), while Zn is again similar to 0.4 dex higher than in giants of similar [Fe/H] (6 stars only). Conclusions. For C, the dwarf/giant discrepancy could possibly have an astrophysical cause, but for the other elements it must arise from shortcomings in the analysis. Approximate computations of granulation (3D) effects yield smaller corrections for giants than for dwarfs, but suggest that this is an unlikely explanation, except perhaps for C, Cr, and Mn. NLTE computations for Na and Al provide consistent abundances between dwarfs and giants, unlike the LTE results, and would be highly desirable for the other discrepant elements as well. Meanwhile, we recommend using the giant abundances as reference data for Galactic chemical evolution models.

Characterization and kinetics of sulfide-oxidizing autotrophic denitrification in batch reactors containing suspended and immobilized cells

Sulfide-oxidizing autotrophic denitrification is an advantageous alternative over heterotrophic denitrification, and may have potential for nitrogen removal of low-strength wastewaters, such as anaerobically pre-treated domestic sewage. This study evaluated the fundamentals and kinetics of this process in batch reactors containing suspended and immobilized cells. Batch tests were performed for different NO(x)(-)/S(2-) ratios and using nitrate and nitrite as electron acceptors. Autotrophic denitrification was observed for both electron acceptors, and NO(x)(-)/S(2-) ratios defined whether sulfide oxidation was complete or not. Kinetic parameter values obtained for nitrate were higher than for nitrite as electron acceptor. Zero-order models were better adjusted to profiles obtained for suspended cell reactors, whereas first-order models were more adequate for immobilized cell reactors. However, in the latter, mass transfer physical phenomena had a significant effect on kinetics based on biochemical reactions. Results showed that sulfide-oxidizing autotrophic denitrification can be successfully established for low-strength wastewaters and have potential for nitrogen removal from anaerobically pre-treated domestic sewage.

Optimal design of periodic functionally graded composites with prescribed properties

The computational design of a composite where the properties of its constituents change gradually within a unit cell can be successfully achieved by means of a material design method that combines topology optimization with homogenization. This is an iterative numerical method, which leads to changes in the composite material unit cell until desired properties (or performance) are obtained. Such method has been applied to several types of materials in the last few years. In this work, the objective is to extend the material design method to obtain functionally graded material architectures, i.e. materials that are graded at the local level (e.g. microstructural level). Consistent with this goal, a continuum distribution of the design variable inside the finite element domain is considered to represent a fully continuous material variation during the design process. Thus the topology optimization naturally leads to a smoothly graded material system. To illustrate the theoretical and numerical approaches, numerical examples are provided. The homogenization method is verified by considering one-dimensional material gradation profiles for which analytical solutions for the effective elastic properties are available. The verification of the homogenization method is extended to two dimensions considering a trigonometric material gradation, and a material variation with discontinuous derivatives. These are also used as benchmark examples to verify the optimization method for functionally graded material cell design. Finally the influence of material gradation on extreme materials is investigated, which includes materials with near-zero shear modulus, and materials with negative Poisson`s ratio.

The generalized log-gamma mixture model with covariates: local influence and residual analysis

In a sample of censored survival times, the presence of an immune proportion of individuals who are not subject to death, failure or relapse, may be indicated by a relatively high number of individuals with large censored survival times. In this paper the generalized log-gamma model is modified for the possibility that long-term survivors may be present in the data. The model attempts to separately estimate the effects of covariates on the surviving fraction, that is, the proportion of the population for which the event never occurs. The logistic function is used for the regression model of the surviving fraction. Inference for the model parameters is considered via maximum likelihood. Some influence methods, such as the local influence and total local influence of an individual are derived, analyzed and discussed. Finally, a data set from the medical area is analyzed under the log-gamma generalized mixture model. A residual analysis is performed in order to select an appropriate model.

A chemometric study on the analgesic activity of cannabinoid compounds using SDA, KNN and SIMCA methods

The supervised pattern recognition methods K-Nearest Neighbors (KNN), stepwise discriminant analysis (SDA), and soft independent modelling of class analogy (SIMCA) were employed in this work with the aim to investigate the relationship between the molecular structure of 27 cannabinoid compounds and their analgesic activity. Previous analyses using two unsupervised pattern recognition methods (PCA-principal component analysis and HCA-hierarchical cluster analysis) were performed and five descriptors were selected as the most relevants for the analgesic activity of the compounds studied: R (3) (charge density on substituent at position C(3)), Q (1) (charge on atom C(1)), A (surface area), log P (logarithm of the partition coefficient) and MR (molecular refractivity). The supervised pattern recognition methods (SDA, KNN, and SIMCA) were employed in order to construct a reliable model that can be able to predict the analgesic activity of new cannabinoid compounds and to validate our previous study. The results obtained using the SDA, KNN, and SIMCA methods agree perfectly with our previous model. Comparing the SDA, KNN, and SIMCA results with the PCA and HCA ones we could notice that all multivariate statistical methods classified the cannabinoid compounds studied in three groups exactly in the same way: active, moderately active, and inactive.

The association between air pollution and blood pressure in traffic controllers in Santo Andre, Sao Paulo, Brazil

Background: Urban air pollutants are associated with cardiovascular events. Traffic controllers are at high risk for pollution exposure during outdoor work shifts. Objective: The purpose of this study was to evaluate the relationship between air pollution and systemic blood pressure in traffic controllers during their work shifts. Methods: This cross-sectional study enrolled 19 male traffic controllers from Santo Andre city (Sao Paulo, Brazil) who were 30-60 years old and exposed to ambient air during outdoor work shifts. Systolic and diastolic blood pressure readings were measured every 15 min by an Ambulatory Arterial Blood Pressure Monitoring device. Hourly measurements (lags of 0-5 h) and the moving averages (2-5 h) of particulate matter (PM(10)), ozone (O(3)) ambient concentrations and the acquired daily minimum temperature and humidity means from the Sao Paulo State Environmental Agency were correlated with both systolic and diastolic blood pressures. Statistical methods included descriptive analysis and linear mixed effect models adjusted for temperature, humidity, work periods and time of day. Results: Interquartile increases of PM(10) (33 mu g/m(3)) and O(3) (49 mu g/m(3)) levels were associated with increases in all arterial pressure parameters, ranging from 1.06 to 2.53 mmHg. PM(10) concentration was associated with early effects (lag 0), mainly on systolic blood pressure. However, O(3) was weakly associated most consistently with diastolic blood pressure and with late cumulative effects. Conclusions: Santo Andre traffic controllers presented higher blood pressure readings while working their outdoor shifts during periods of exposure to ambient pollutant fluctuations. However, PM(10) and O(3) induced cardiovascular effects demonstrated different time courses and end-point behaviors and probably acted through different mechanisms. (C) 2011 Elsevier Inc. All rights reserved.

Postmaterialist and economic voting in Australia, 1990-98

Recent research in Australian sociology and political science has debated the extent to which postmaterialist values and economic self-interest shape voting in federal elections. Some researchers have argued that postmaterialist values have partly displaced materialist concerns with physical security and economic well-being in Australian public life. This displacement, coupled with the adoption by major political parties of postmaterialist 'quality of life' issues such as the environment, has meant that voting in Australia has come to be more dependent on postmaterialist values than on perceptions of economic interest. Other research, however, has found no relationship between postmaterialist values and voting behaviour, while economic evaluations remain a strong determinant of voting behaviour. Part of the disagreement reflects methodological differences in the research. But different methodological problems compromise each of the previous studies. In this paper we use data from the 1990, 1993, 1996 and 1998 Australian Election Studies to investigate postmaterialist and economic voting in the Commonwealth House of Representatives and the Senate. Using various statistical methods, we first explore bivariate relationships between key variables and then use multivariate models of postmaterialist and economic voting to adjudicate between the contending positions.

The application of statistical process control charts to the detection and monitoring of hospital-acquired infections

The monitoring of infection control indicators including hospital-acquired infections is an established part of quality maintenance programmes in many health-care facilities. However, surveillance data use can be frustrated by the infrequent nature of many infections. Traditional methods of analysis often provide delayed identification of increasing infection occurrence, placing patients at preventable risk. The application of Shewhart, Cumulative Sum (CUSUM) and Exponentially Weighted Moving Average (EWMA) statistical process control charts to the monitoring of indicator infections allows continuous real-time assessment. The Shewhart chart will detect large changes, while CUSUM and EWMA methods are more suited to recognition of small to moderate sustained change. When used together, Shewhart and EWMA methods are ideal for monitoring bacteraemia and multiresistant organism rates. Shewhart and CUSUM charts are suitable for surgical infection surveillance.

PAH air pollution at a Portuguese urban area: carcinogenic risks and sources identification

This study aimed to characterize air pollution and the associated carcinogenic risks of polycyclic aromatic hydrocarbon (PAHs) at an urban site, to identify possible emission sources of PAHs using several statistical methodologies, and to analyze the influence of other air pollutants and meteorological variables on PAH concentrations.The air quality and meteorological data were collected in Oporto, the second largest city of Portugal. Eighteen PAHs (the 16 PAHs considered by United States Environment Protection Agency (USEPA) as priority pollutants, dibenzo[a,l]pyrene, and benzo[j]fluoranthene) were collected daily for 24 h in air (gas phase and in particles) during 40 consecutive days in November and December 2008 by constant low-flow samplers and using polytetrafluoroethylene (PTFE) membrane filters for particulate (PM10 and PM2.5 bound) PAHs and pre-cleaned polyurethane foam plugs for gaseous compounds. The other monitored air pollutants were SO2, PM10, NO2, CO, and O3; the meteorological variables were temperature, relative humidity, wind speed, total precipitation, and solar radiation. Benzo[a]pyrene reached a mean concentration of 2.02 ngm−3, surpassing the EU annual limit value. The target carcinogenic risks were equal than the health-based guideline level set by USEPA (10−6) at the studied site, with the cancer risks of eight PAHs reaching senior levels of 9.98×10−7 in PM10 and 1.06×10−6 in air. The applied statistical methods, correlation matrix, cluster analysis, and principal component analysis, were in agreement in the grouping of the PAHs. The groups were formed according to their chemical structure (number of rings), phase distribution, and emission sources. PAH diagnostic ratios were also calculated to evaluate the main emission sources. Diesel vehicular emissions were the major source of PAHs at the studied site. Besides that source, emissions from residential heating and oil refinery were identified to contribute to PAH levels at the respective area. Additionally, principal component regression indicated that SO2, NO2, PM10, CO, and solar radiation had positive correlation with PAHs concentrations, while O3, temperature, relative humidity, and wind speed were negatively correlated.

Modelos de previsão de produção para centrais mini-hídricas

O constante crescimento dos produtores em regime especial aliado à descentralização dos pontos injetores na rede, tem permitido uma redução da importação de energia mas também tem acarretado maiores problemas para a gestão da rede. Estes problemas estão relacionados com o facto da produção estar dependente das condições climatéricas, como é o caso dos produtores eólicos, hídricos e solares. A previsão da energia produzida em função da previsão das condições climatéricas tem sido alvo de atenção por parte da comunidade empresarial do setor, pelo facto de existir modelos razoáveis para a previsão das condições climatéricas a curto prazo, e até a longo prazo. Este trabalho trata, em concreto, do problema da previsão de produção em centrais mini-hídricas, apresentando duas propostas para essa previsão. Em ambas as propostas efetua-se inicialmente a previsão do caudal que chega à central, sendo esta depois convertida em potência que é injetada na rede. Para a previsão do caudal utilizaram-se dois métodos estatísticos: o método Holt-Winters e os modelos ARMAX. Os dois modelos de previsão propostos consideram um horizonte temporal de uma semana, com discretização horária, para uma central no norte de Portugal, designadamente a central de Penide. O trabalho também contempla um pequeno estudo da bibliografia existente tanto para a previsão da produção como de afluências de centrais hidroelétricas. Aborda, ainda, conceitos relacionados com as mini-hídricas e apresenta uma caraterização do parque de centrais mini-hídricas em Portugal.

Teorias e modelos organizacionais

Este trabalho propõe-se a investigar as teorias e modelos organizacionais e a respetiva aplicabilidade nas organizações portuguesas. Após a revisão da literatura sobre modelos organizacionais, foi efetuada uma investigação quantitativa através de um questionário online com a finalidade de avaliar quais os modelos organizacionais predominantemente utilizados e quais as características organizacionais que levam à utilização de determinado modelo. Através de métodos estatísticos analisaram-se os resultados do inquérito com o objetivo de verificar a existência de possíveis relações entre diversas características das organizações e o modelo organizacional usado. Foi possível concluir que o modelo organizacional Burocrático é o modelo predominantemente utilizado pelos respondentes e que as organizações que adotam o modelo burocrático parecem conseguir implementar processos sistemáticos de inovação compatibilizando as regras e procedimentos com a capacidade para aprender e se adaptar. O setor de atividade e a dimensão das organizações são as variáveis que mais influenciam a adoção do modelo organizacional. A investigação contribui para o conhecimento teórico e pratico sobre modelos organizacionais e sobre a sua aplicação em diferentes tipos de organizações portuguesas e para a compreensão e capacitação dos engenheiros do tema da cultura organizacional, de modo a poderem trabalhar de forma efetiva em grupos multidisciplinares que criem valor para as respetivas organizações, inovando e aplicando a engenharia e tecnologia para lidar com as questões e desafios atuais referidos pelo relatório da UNESCO (1).

Modeling in-Hospital Patient Survival During the First 28 Days After Intensive Care Unit Admission: a Prognostic Model for Clinical Trials in General Critically Ill Patients

OBJECTIVE: The objective of the study was to develop a model for estimating patient 28-day in-hospital mortality using 2 different statistical approaches. DESIGN: The study was designed to develop an outcome prediction model for 28-day in-hospital mortality using (a) logistic regression with random effects and (b) a multilevel Cox proportional hazards model. SETTING: The study involved 305 intensive care units (ICUs) from the basic Simplified Acute Physiology Score (SAPS) 3 cohort. PATIENTS AND PARTICIPANTS: Patients (n = 17138) were from the SAPS 3 database with follow-up data pertaining to the first 28 days in hospital after ICU admission. INTERVENTIONS: None. MEASUREMENTS AND RESULTS: The database was divided randomly into 5 roughly equal-sized parts (at the ICU level). It was thus possible to run the model-building procedure 5 times, each time taking four fifths of the sample as a development set and the remaining fifth as the validation set. At 28 days after ICU admission, 19.98% of the patients were still in the hospital. Because of the different sampling space and outcome variables, both models presented a better fit in this sample than did the SAPS 3 admission score calibrated to vital status at hospital discharge, both on the general population and in major subgroups. CONCLUSIONS: Both statistical methods can be used to model the 28-day in-hospital mortality better than the SAPS 3 admission model. However, because the logistic regression approach is specifically designed to forecast 28-day mortality, and given the high uncertainty associated with the assumption of the proportionality of risks in the Cox model, the logistic regression approach proved to be superior.

Avaliação do desempenho de fundos de investimento de obrigações: evidência para o mercado Brasileiro

Dissertação de mestrado em Finanças