Using data envelopment analysis with principal components, we calculate an adoption-intensity index for herbicide-resistance best management practices (BMPs). Empirical results for over 1,100 farmers in twenty-two U.S. states suggest that many farmers could improve their herbicide resistance BMP adoption. Two-limit truncated regression results show that higher yields and a greater proportion of acres planted with Roundup Ready (R) seeds motivate weed BMP adoption. While soybean and corn farmers have lower adoption intensity than cotton farmers, farmer educational attainment and greater concern for herbicide effectiveness and for human and environmental safety are found to help increase the adoption of weed BMPs.

Background: Antimicrobial-resistant infections result in hospital stays costing between $18,000 and $29,000. As of 2009, Centers for Medicare and Medicaid Services no longer upgrade payments for hospital-acquired infections. Hospital epidemiologists monitor and document rates of individual resistant microbes in antibiogram reports. Overall summary measures capturing resistance within a hospital may be useful. Objectives: We applied four techniques (L1- and L2-principal component analysis (PCA), desirability functions, and simple summary) to create summary measures of resistance and described the four summary measures with respect to reliability, proportion of variance explained, and clinical utility. Methods: We requested antibiograms from hospitals participating in the University HealthSystem Consortium for the years 2002–2008 (n=40). A clinical team selected organism-drug resistant pairs (as resistant isolates per 1,000 patient days) based on 1) virulence, 2) complicated or toxic therapies, 3) transmissibility, and 4) high incidence with increasing levels of resistance. Four methods were used to create summary scores: 1) L1- and L2-PCA: derived multipliers so that the variance explained is maximized; 2) desirability function: transformed resistance data to be between 0 and 1; 3) simple sum: each resistance rate was added and divided by the square root of the total number of microbes summed. Simple correlation analyses between time and each summary score evaluated reliability. For each year, we calculated the proportion of explained variance by dividing each summary score’s variance by the variance in the original data. Clinical utility was checked by comparing the trends for all of the individual microbe’s resistance rates to the trends seen in the summary scores for each hospital. Results: Proportion of variance explained by L1- and L2-PCA and the simple sum was 0.61, 0.62, and 0.29 respectively. Simple sum and L1- and L2-PCA summary scores best followed the trends seen in the individual antimicrobial resistance rates; trends in desirability function scores deviated from those seen in individual trends of antimicrobial resistance. L1- and L2-PCA summary scores were more influenced by MRSA rates, and the simple sum score was less influenced. Pearson correlation coefficients revealed good reliability through time. Conclusion: Deriving summary measures of antimicrobial resistance can be reliable over time and explain a high proportion of variance. Infection control practitioners and hospital epidemiologists may find the inclusion of a summary score of antimicrobial resistance beneficial in describing the trends of overall resistance in their yearly antibiogram reports.

Community ecology from the perspective of classic and Bayesian statistics Ekologie společenstev z hlediska klasické a Bayesovské statistiky Řešitel: Adam Klimeš Vedoucí práce: Mgr. Petr Keil, Ph.D. Abstract Quantitative evaluation of evidence through statistics is a central part of present-day science. Bayesian approach represents an emerging but rapidly developing enrichment of statistical analysis. The approach differs in its foundations from the classic methods. These differences, such as the different interpretation of probability, are often seen as obstacles for acceptance of Bayesian approach. In this thesis I outline ways to deal with the assumptions of Bayesian approach, and I address the main objections against it. I present Bayesian approach as a new way to handle data to answer scientific questions. I do this from a standpoint of community ecology: I illustrate the novelty that Bayesian approach brings to data analysis of typical community ecology data, specifically, the analysis of multivariate datasets. I focus on principal component analysis, one of the typical and frequently used analytical techniques. I execute Bayesian analyses that are analogical to the classic principal components analysis, I report the advantages of the Bayesian version, such as the possibility of working with...

Plus de 150 millions de personnes souffrent de troubles dépressifs à travers le monde. Malgré le nombre croissant d’études s’intéressant à la physiopathologie de ce trouble, aucune amélioration majeure concernant ses traitements ou la compréhension des mécanismes biologiques sous-jacents n’a été faite. Bien qu’une prédisposition génétique et des évènements stressants aient été proposés comme facteurs de risque, ni les gènes impliqués ni le fonctionnement des interactions gène x environnement ne sont encore connus. Cela peut s’expliquer par le manque de modèles animaux satisfaisants et par le fossé existant entre les connaissances / méthodes de diagnostic appliquées en recherche clinique et celles disponibles en recherche fondamentale. Des manipulations pharmacologiques, lésionnelles, génétiques ou de l’environnement sont quasi exclusivement utilisées chez le rongeur. Certains primates non-humains (PNH), plus proches de nous sur les plans comportementaux et phylogénétiques, montrent pourtant, comme l’Homme, des modifications comportementales et physiologiques atypiques et spontanées en réponse à des conditions de vie stressantes. Malgré les travaux pionniers et prometteurs d’Harlow et de ses collaborateurs dans les années 60, rares sont les équipes qui étudient la dépression chez le macaque aujourd’hui. Nous avons émis l’hypothèse que parmi des grandes populations de PNH captifs, une petite proportion d’individus exprime des comportements atypiques pouvant s’apparenter à des symptômes dépressifs. Mon projet de thèse a eu pour but de proposer une approche novatrice et non invasive d’identification de ces profils « depressive-like » chez le macaque, en combinant les compétences et connaissances de l’éthologie, de la psychiatrie et des neurosciences. L’impact des expériences de vie précoces et de l’espèce a également été abordé. Les comportements, les postures et orientations du corps, les localisations spatiales, les regards et/ou les distances inter-individuelles ont été relevés chez plus de 200 macaques rhésus et cynomolgus d’élevage, nés en captivité ou dans la nature. Des sous-groupes d’individus ont été identifiés à l’aide d’analyses multifactorielles. Dans chaque population observée, un profil « depressive-like » a été mis en évidence par comparaison avec les symptômes décrits dans le Manuel Diagnostique et Statistiques des Troubles Mentaux et avec les modèles animaux existants dans la littérature. La prévalence de ces profils étant supérieure chez les macaques rhésus et chez les animaux nés en captivité, nos résultats concordent avec le rôle suggéré du stress dans l’expression des troubles dépressifs. En plus d’exprimer ce profil comportemental atypique dans leur environnement habituel, les singes « depressive-like » présentaient une réactivité émotionnelle altérée au cours 2 tests comportementaux, associée à des taux élevés de cortisol plasmatique et noradrénaline cérébro-spinale. Pris dans leur ensemble, ces résultats prometteurs confèrent une bonne validité de représentation à notre modèle macaque de symptômes dépressifs. Une caractérisation plus complète de ce modèle est bien sûr nécessaire et pourrait ouvrir de nouvelles perspectives quant à la compréhension de l’étiologie et de la physiopathologie des troubles dépressifs. / More than 150 million people worldwide suffer from major depressive disorder (MDD). Although investigations of its pathophysiology have dramatically increased in the last decade, no substantial improvement has been made concerning the treatments and the understanding of its underlying mechanisms. A genetic predisposition and stressful experiences have been acknowledged as risk factors involved in MDD. However, no specific genes have been identified so far and little is known about the gene x environment interactions. This is likely due to the lack of bona fide animal models of depressive-like symptoms. Indeed, there is a huge gap between the knowledge / diagnostic methodology of clinical research and the animal models used in fundamental research, mainly focusing on environmental, pharmacological, lesional or genetic manipulations. Phylogenetically and behaviourally closer to Humans compared to rodents, non-human primates (NHPs) can show spontaneous behavioural and physiological modifications in response to stressful life events. Although promising results had been reported in the 1960’s by the pioneering studies of Harlow and colleagues, the investigation of depressive-like symptoms in macaques are scarce in the current literature. We hypothesize that, among large captive NHP populations, a few individuals will display atypical behaviours that could mimic depressive symptoms. Combining the skills and knowledge of ethology, psychiatry and neurosciences, my PhD project aimed at proposing an innovative non-invasive detection method of such depressive-like profiles. The impact of birth origin and species was questioned as well. Behaviours, body postures, body orientations, spatial location, gaze direction and/or inter-peer distances were collected among more than 200 rhesus and cynomolgus captive- or wild-born farm-bred macaques. Using multifactorial analyses, clusters of individuals displaying distinct behavioural profiles were identified. In each population, a common depressive-like profile was characterised by its similarities with symptoms described in the Diagnostic and Statistical Manual of Mental Disorder and with other animal models of depression. The prevalence of such profiles was increased in the rhesus populations and by captive early life experience, corroborating the role of stress in the development of MDD. In addition to expressing depressive-like features in their home cage, these animals displayed higher levels of plasmatic cortisol and cerebrospinal noradrenaline which correlated with a passive emotional reactivity in 2 behavioural paradigms. Altogether these promising results conferred good face validity to our NHP model of depressive-like symptoms. Further characterization of this model is required and might bring new insights to the understanding of MDD pathophysiology and etiology.

The first chapter investigates how corporate governance influences firms’ capital structure behavior. Based on the premise that costs associated with deviations from the target capital structure are positively correlated to the extent of deviation, we hypothesize that the initial deviation from the target will be shorter for a firm with good corporate governance than for a firm with poor corporate governance. We also hypothesize that the former group will employ a higher speed of adjustment towards target than the latter group due primarily to the following reasons. First, a firm with well-placed governance system will adjust at a faster rate because longer it stays deviated, the higher the loss of value it faces. Second, firms with better governance structures enjoy lower adjustment costs. We develop three sets of measures for the quality of corporate governance and analyze how they influence a firm’s rebalancing behavior in presence of relevant control variables. Our results are consistent with the hypotheses. The second chapter explores investors’ reactions to new information on REITs and non-REITs option markets. The real estate market can be fairly volatile; what remains unclear is whether price changes are excessively volatile relative to fundamentals. This study attempts to examine the latter by using the methodology based on Stein (1989), which utilizes option data. The advantage of using option data rather than stock data to assess the reactions to information is that option valuation is not affected by changes in risk premium. Under volatility mean reversion, the changes in implied volatilities of long-term options should be less than those of short-term options. If not, an excessive reaction is suggested. Specifically, the study compares the changes in implied volatilities of options on REITs and non-REITs. Because real estate transactions typically involve a great degree of leverage, reactions can be greater for REITs than for non-REITs; on the other hand, there are several reasons that REITs are subject to potentially a lower degree of excessive reactions. Empirical results indicate that the reactions to information are stronger in non-REITs than in REITs. Moreover, we find that down markets are associated with stronger reactions, which we argue might be due to a leverage effect.

[pt] O desenvolvimento de um reservatório de petróleo já conhecido e delimitado consiste em encontrar uma alternativa (configuração) de poços que contribua para maximizar a receita a ser obtida com o óleo recuperado do reservatório. A busca por esta alternativa frequentemente é baseada em processos de otimização que usam o valor presente líquido (VPL) do projeto como função de avaliação das alternativas encontradas durante a busca. Dentre outras variáveis, o cálculo do VPL é diretamente dependente dos dados de produção de óleo, gás e água durante a vida produtiva do reservatório, bem como de seus custos de desenvolvimento. Determinar a localização, os tipos (produtor ou injetor) e a trajetória de poços em um reservatório é um problema de otimização complexo que depende de uma grande quantidade de variáveis, dentre elas as propriedades do reservatório (tais como porosidade e permeabilidade) e os critérios econômicos. Os processos de otimização aplicados a este tipo de problema têm um alto custo computacional devido ao uso contínuo de simuladores que reproduzem as condições do reservatório e do sistema de superfície. O uso dos simuladores pode ser substituído por um aproximador, que neste trabalho, é um modelo que utiliza Redes Neurais Artificiais. Os aproximadores aqui apresentados são feitos para substituir a simulação integrada do reservatório, do poço e da superfície (linhas de produção e riser). As amostras para a construção do aproximador é feita utilizando os simuladores de reservatório e de superfície e para reduzir o número de amostras necessárias e tornar sua construção mais rápida, utiliza-se Hipercubo Latino e Análise de Componentes Principais. Os aproximadores foram testados em dois reservatórios petrolíferos: um reservatório sintético, e baseado em um caso real. Os resultados encontrados indicam que estes aproximadores conseguem bom desempenho na substituição dos simuladores no processo de otimização devido aos baixos erros encontrados e à substancial diminuição do custo computacional. / [en] The development of an oil reservoir consists in finding an alternative of wells that contributes to maximizing the revenue to be obtained from the recovered reservoir oil. The pursuit for this alternative is often based on optimization processes using the net present value (NPV) of the project as the evaluation function of the alternatives found during this pursuit. Among other variables, the NPV calculation is directly dependent on the oil, gas and water production data during the productive life of the reservoir, as well as their development costs. Determine the number, location, type (producer or injector) and the trajectory of wells in a reservoir is a complex optimization problem which depends on a lot of variables, including the reservoir properties (such as porosity and permeability) and economic criteria. The optimization processes applied to this type of problem has a high computational cost due to the continuous use of simulators that reproduce the conditions of the reservoir and the surface system. The use of simulators may be replaced by proxies. At the present work, proxies were constructed using artificial neural networks. The proxies presented here are meant to replace the integrated reservoir, well and surface (production lines and riser) simulation to reduce the computational cost of a decision support system. The samples for the construction of the proxies are produced using reservoir and surface simulators. To reduce the number of samples needed for the proxy construction, and, to reduce the dimension of the problem, Latin Hypercube and Principal Component Analysis are used. The approximators were tested in two oil reservoirs: a synthetic reservoir, and another with real features. The results indicate that these approximators can perform well in replacement of simulators in the optimization process due to low errors found and a substantial decrease in computational cost.

Grâce aux informations spatiales et spectrales qu'elle apporte, l'imagerie multispectrale de la peau est devenue un outil incontournable de la dermatologie. Cette thèse a pour objectif d'évaluer l’intérêt de cet outil pour la cosmétologie à travers trois études : la détection d'un fond de teint, l'évaluation de l'âge et la mesure de la rugosité.Une base d'images multispectrales de peau est construite à l'aide d'un système à multiples filtres optiques. Une phase de prétraitement est nécessaire à la standardisation et à la mise en valeur de la texture des images.Les matrices de covariance des acquisitions peuvent être représentées dans un espace multidimensionnel, ce qui constitue une nouvelle approche de visualisation de données multivariées. De même, une nouvelle alternative de réduction de la dimensionnalité basée sur l'ACP est proposée dans cette thèse. L'analyse approfondie de la texture des images multispectrales est réalisée : les paramètres de texture issus de la morphologie mathématique et plus généralement de l'analyse d'images sont adaptés aux images multivariées. Dans cette adaptation, plusieurs distances spectrales sont expérimentées, dont une distance intégrant le modèle LIP et la métrique d'Asplünd.Les résultats des prédictions statistiques générées à partir des données de texture permettent de conclure quant à la pertinence du traitement des données et de l'utilisation de l'imagerie multispectrale pour les trois études considérées. / Thanks to its precision in spatial and spectral domain, multispectral imaging has become an essential tool in dermatology. This thesis focuses on the interest of this technology for cosmetological parameters assessment through three different studies: the detection of a foundation make-up, age assessment and roughness measurement.A database of multispectral skin images is build using a multiple optical filters system. A preprocessing step allows to standardize those texture images before their exploitation.Covariance matrices of mutispectral acquisitions can be displayed in a multidimensional scaling space which is a novel way to represent multivariate data sets. Likewise, a new dimensionality reduction algorithm based on PCA is proposed in this thesis.A complete study of the images texture is performed: texture features from mathematical morphology and more generally from image analysis are expanded to the case of multivariate images. In this process, several spectral distances are tested, among which a new distance associating the LIP model to the Asplund metric.Statistical predictions are generated from texture data. Thoses predictions lead to a conclusion about the data processing efficiency and the relevance of multispectral imaging for the three cosmetologic studies.

No description available.

As cisteíno-proteases da família da papaína desempenham funções essenciais em processos biológicos, entre eles o desenvolvimento e crescimento do organismo, vias de sinalização celular e apoptose, invasão de parasitas em células hospedeiras. Assim, trata-se de uma classe de proteínas de grande interesse para as indústrias farmacêuticas, sendo utilizada como alvo para o tratamento de doenças como o câncer e metástases, osteoporose. Disfunções relacionadas ao sistema imune, doenças parasitárias como malária, leishmaniose, doença do sono e doença de Chagas. Esta última é uma enfermidade considerada negligenciada pelas grandes indústrias farmacêuticas, sem nenhum tratamento eficaz e seguro disponível, que gera um problema econômico de mais de sete bilhões de dólares anuais devido à perda de mão de obra e gastos com tratamento para amenizar os efeitos da doença. A cisteíno protease cruzaína de Trypanosoma cruzi, causador da Doença de Chagas, desponta como um alvo validado na busca de novos fármacos contra essa enfermidade. Essa enzima apresenta um par de aspartatos que interagem entre si, para os quais foi predito um pKa de 7, sendo possível a forma desprotonada desse par em condições biológicas. Neste caso, pode levar à exposição de uma nova cavidade por meio do movimento da alça entre os resíduos 57-62, segundo as simulações de dinâmica molecular desse trabalho, que se trata de uma possível candidata a ponto de seletividade de inibidores de cisteínoproteases de parasitos em relação às suas ortólogas em Homo sapiens que não possuem o par de aspartatos. Em pH ácido, foi mostrado por meio de análise de componentes principais de simulações de dinâmica molecular que as cisteíno protease apresentam uma restrição gradual na amostragem conformacional do sítio ativo quando complexadas com as formas não covalente e covalente de inibidores derivados de dipeptidil nitrilas. Isso sugere que esse sistema segue o modelo de seleção conformacional para flexibilidade de proteína. Notou-se também que o perfil de restrição de ligantes que inibem na faixa de nmol.L-1 difere daqueles a µmol.L-1 , o que possibilitou a construção de uma árvore de decisão para identificar os complexos que apresentam afinidade a nmol.L-1 . / The papain-like cysteine proteases are essentials for biological process, performing important roles on the parasite development, growth and also in the parasite invasion process on the host cell, in cellular signaling pathways and apoptosis, among others. Thus, the pharmaceutical industry widely uses this class of protein as target for the development of new drugs, against cancer and metastasis, osteoporosis and immune system disorders, resulting in many approved drugs. Additionally, these enzymes are validated target against parasitic diseases as leishmaniose, malaria and African and American trypanosomiasis. The last one, also known as Chagas\' disease, is neglected disease for which, further a century form this discovery, there is no effective and safe chemotherapy and is responsible for an economic loss of around seven billion dollars in the world per year due to the health care and lost productivity from infected people. Faced with this situation, the Cruzain, a cysteine protease from the Trypanosoma cruzi, the causative agent of Chagas\' disease, is emerging as interesting and validated target to the search for new drugs against this sickness. This enzyme has a pair of interactiong Asp for which was predicted a pKa of 7, by computational methods. By this way, this pair under neutral to alkaline pH adopts the deprotonated form, which exposed a new cavity through a movement of loop of residues 57-62, as we showed here by molecular dynamics simulations. This cavity emerges as a possible selectivity point of the cruzain inhibitors, once Homo sapiens enzynes does not present the aspartic acid - aspartate pair. In condition of acidic pH, principal component analysis of molecular dynamics simulations showed a gradual reduction of the conformational space covered by the active site of cruzain, cathepsin K and cathepsin L in it free form and complexed with dipeptidyl nitrilelike molecules in it noncovalent e covalent forms. This suggests these systems follows the conformational selection model of protein flexibility. Furthermore, we observed the ligands that inhibits the protein at nmol.L-1 induces the protein flexibility in a similar way, while the µmol.L-1 ones leads to another pattern. That made possible the construction of a decision tree which is able to identify nmol.L-1 from µmol.L-1 complexes.

Accurate information about tree volume in tropical vegetation formations is critical for the identification of potential areas for sustainable timber production, carbon estimation and biodiversity conservation. Difficult access and the cost of obtaining a large number of samples needed for accurate wood volume and biomass determination are often barriers for carrying out inventories and studies in natural forests in Brazil. Therefore, the development of more efficient techniques of mensuration in tropical forests is an important mechanism for conservation, management and production advancement. The main purposes of this thesis are: introducing a new method for quantifying vertical and horizontal structures by using principal component analysis (PCA); developing two different approaches of volume modeling, one based on DBH and another based on crown area; proposing a new taper equation for native vegetation in three different formations; and estimating upper section diameters to become the geometric form method useful in natural vegetation in Brazil and reducing dependence on destructive measurements. / Informações corretas sobre o volume de árvores em formações de vegetação natural são fundamentais para a identificação de áreas potenciais para produção madeireira sustentável, estimativa de carbono e conservação da biodiversidade. Dificuldade de acesso e altos custos na obtenção de amostras necessárias para estimativas precisas de volume e biomassa são barreiras frequentes na condução de inventários e estudos florestais no Brasil. Dessa forma, o desenvolvimento de técnicas mais eficientes de mensuração em florestas tropicais é um importante mecanismo para o avanço da conservação, manejo e produção. Os principais objetivos deste trabalho foram: introduzir um novo método de quantificação das estruturas vertical e horizontal por meio do uso de análise de componentes principais (ACP); desenvolver modelos volumétricos baseados em DAP e modelos baseados em área de copa; propor uma nova função de afilamento aplicada a três diferentes formações vegetais; e estimar alturas em que deverão ter diâmetros medidos ao longo da árvore para tornar o método geométrico útil em formações naturais no Brasil, e reduzindo a necessidade de mensurações que requerem o abatimento das árvores.