Conditioning of the weak-constraint variational data assimilation problem for numerical weather prediction

4-Dimensional Variational Data Assimilation (4DVAR) assimilates observations through the minimisation of a least-squares objective function, which is constrained by the model flow. We refer to 4DVAR as strong-constraint 4DVAR (sc4DVAR) in this thesis as it assumes the model is perfect. Relaxing this assumption gives rise to weak-constraint 4DVAR (wc4DVAR), leading to a different minimisation problem with more degrees of freedom. We consider two wc4DVAR formulations in this thesis, the model error formulation and state estimation formulation. The 4DVAR objective function is traditionally solved using gradient-based iterative methods. The principle method used in Numerical Weather Prediction today is the Gauss-Newton approach. This method introduces a linearised `inner-loop' objective function, which upon convergence, updates the solution of the non-linear `outer-loop' objective function. This requires many evaluations of the objective function and its gradient, which emphasises the importance of the Hessian. The eigenvalues and eigenvectors of the Hessian provide insight into the degree of convexity of the objective function, while also indicating the difficulty one may encounter while iterative solving 4DVAR. The condition number of the Hessian is an appropriate measure for the sensitivity of the problem to input data. The condition number can also indicate the rate of convergence and solution accuracy of the minimisation algorithm. This thesis investigates the sensitivity of the solution process minimising both wc4DVAR objective functions to the internal assimilation parameters composing the problem. We gain insight into these sensitivities by bounding the condition number of the Hessians of both objective functions. We also precondition the model error objective function and show improved convergence. We show that both formulations' sensitivities are related to error variance balance, assimilation window length and correlation length-scales using the bounds. We further demonstrate this through numerical experiments on the condition number and data assimilation experiments using linear and non-linear chaotic toy models.

Implicit equal-weights particle filter

Filter degeneracy is the main obstacle for the implementation of particle filter in non-linear high-dimensional models. A new scheme, the implicit equal-weights particle filter (IEWPF), is introduced. In this scheme samples are drawn implicitly from proposal densities with a different covariance for each particle, such that all particle weights are equal by construction. We test and explore the properties of the new scheme using a 1,000-dimensional simple linear model, and the 1,000-dimensional non-linear Lorenz96 model, and compare the performance of the scheme to a Local Ensemble Kalman Filter. The experiments show that the new scheme can easily be implemented in high-dimensional systems and is never degenerate, with good convergence properties in both systems.

Breaking the double-edged sword of effort/trying hard: developmental equilibrium and Longitudinal relations among effort, achievement, and academic self-concept

Ever since the classic research of Nicholls (1976) and others, effort has been recognized as a double-edged sword: whilst it might enhance achievement, it undermines academic self-concept (ASC). However, there has not been a thorough evaluation of the longitudinal reciprocal effects of effort, ASC and achievement,in the context of modern self-concept theory and statistical methodology. Nor have there been developmental equilibrium tests of whether these effects are consistent across the potentially volatile early-to-middle adolescence. Hence, focusing on mathematics, we evaluate reciprocal effects models over the first four years of secondary school, relating effort, achievement (test scores and school grades), ASC, and ASCxEffort interactions for a representative sample of 3,421 German students (Mn age = 11.75 years at Wave 1). ASC, effort and achievement were positively correlated at each wave, and there was a clear pattern of positive reciprocal positive effects among ASC, test scores and school grades—each contributing to the other, after controlling for the prior effects of all others. There was an asymmetrical pattern of effects for effort that is consistent with the double-edged sword premise: prior school grades had positive effects on subsequent effort, but prior effort had non-significant or negative effects on subsequent grades and ASC. However, on the basis of a synergistic application of new theory and methodology, we predicted and found a significant ASC-by-effort interaction, such that prior effort had more positive effects on subsequent ASC and school grades when prior ASC was high—thus providing a key to breaking the double-edged sword.

Toward a resolution of Campanulid phylogeny, with special reference to the placement of Dipsacales

Broad-scale phylogenetic analyses of the angiosperms and of the Asteridae have failed to confidently resolve relationships among the major lineages of the campanulid Asteridae (i.e., the euasterid II of APG II, 2003). To address this problem we assembled presently available sequences for a core set of 50 taxa, representing the diversity of the four largest lineages (Apiales, Aquifoliales, Asterales, Dipsacales) as well as the smaller ""unplaced"" groups (e.g., Bruniaceae, Paracryphiaceae, Columelliaceae). We constructed four data matrices for phylogenetic analysis: a chloroplast coding matrix (atpB, matK, ndhF, rbcL), a chloroplast non-coding matrix (rps16 intron, trnT-F region, trnV-atpE IGS), a combined chloroplast dataset (all seven chloroplast regions), and a combined genome matrix (seven chloroplast regions plus 18S and 26S rDNA). Bayesian analyses of these datasets using mixed substitution models produced often well-resolved and supported trees. Consistent with more weakly supported results from previous studies, our analyses support the monophyly of the four major clades and the relationships among them. Most importantly, Asterales are inferred to be sister to a clade containing Apiales and Dipsacales. Paracryphiaceae is consistently placed sister to the Dipsacales. However, the exact relationships of Bruniaceae, Columelliaceae, and an Escallonia clade depended upon the dataset. Areas of poor resolution in combined analyses may be partly explained by conflict between the coding and non-coding data partitions. We discuss the implications of these results for our understanding of campanulid phylogeny and evolution, paying special attention to how our findings bear on character evolution and biogeography in Dipsacales.

Estimação conjunta das elasticidades: preço da escolha da marca e da quantidade comprada com dados escaneados

O objetivo geral desta tese é estimar a elasticidade-preço da demanda de forma conjunta, decomposta em elasticidade-preço da escolha da marca e da quantidade comprada, e discutir as implicações desta decomposição para uma categoria específica de produto. Para isto foram usados dados escaneados de uma amostra de domicílios no contexto varejista brasileiro. Oito hipóteses foram testadas por meio de dois modelos. O primeiro refere-se à decisão de escolha da marca, em que foi empregado o modelo logit condicional baseado na maximização da utilidade do domicílio. O segundo envolveu equações de demanda, obtidas pelo modelo clássico de regressão linear. Ambos foram especificados de forma que se pudesse testar a dependência das duas decisões de compra. No que diz respeito à validação, o modelo de escolha da marca demonstrou uma satisfatória capacidade de previsão, comparativamente aos modelos analisados na literatura. Implicações gerenciais incluem específicas decisões e ações de preço para as marcas, já que a natureza da decomposição das elasticidades-preço varia entre marcas.

Impactos da política fiscal e tributária no crescimento econômico

Utilizando uma adaptação do modelo de Telles e Mussolini (2014), o presente trabalho busca discutir as caracterizações da matriz tributária, discutindo como as variações na matriz tributária podem impactar na política fiscal. Discute-se com base na diferenciação dos impostos em cinco grupos, a saber: impostos sobre salários, impostos sobre investimentos, impostos sobre consumo, impostos sobre renda e riqueza, e impostos sobre comércio internacional, a partir dos quais argumenta-se quais seus impactos para a política fiscal, para a interação das variáveis econômicas e a evolução destas variáveis, assim como a relação entre a preferência por uma determinada estrutura tributária e o endividamento de um país. Após a discussão teórica, faz-se uma análise descritiva da evolução destas variáveis tributárias para cada categoria de tributo, relacionando a sua evolução no tempo para um grupo de 64 países, tomados a partir do trabalho de Telles e Mussolini (2014), assim como relacionando a sua evolução intertemporal. Por fim, faz-se uma análise da estrutura tributária destes países, discutindo, na análise dos dados em painel, os resultados para as estimativas em modelos de efeitos fixos e efeitos aleatórios, os resultados da estimação pelo modelo Arellano-Bond, e utilizando-se a abordagem instrumental pelo Método Generalizado dos Momentos, onde se conclui que a taxação sobre a riqueza e a taxação sobre o comércio internacional com fins de financiar o excesso de gastos do governo impacta de forma negativa na trajetória de crescimento, entre outros resultados.

Identificação fuzzy-multimodelos para sistemas não lineares

This paper presents a new multi-model technique of dentification in ANFIS for nonlinear systems. In this technique, the structure used is of the fuzzy Takagi-Sugeno of which the consequences are local linear models that represent the system of different points of operation and the precursors are membership functions whose adjustments are realized by the learning phase of the neuro-fuzzy ANFIS technique. The models that represent the system at different points of the operation can be found with linearization techniques like, for example, the Least Squares method that is robust against sounds and of simple application. The fuzzy system is responsible for informing the proportion of each model that should be utilized, using the membership functions. The membership functions can be adjusted by ANFIS with the use of neural network algorithms, like the back propagation error type, in such a way that the models found for each area are correctly interpolated and define an action of each model for possible entries into the system. In multi-models, the definition of action of models is known as metrics and, since this paper is based on ANFIS, it shall be denominated in ANFIS metrics. This way, ANFIS metrics is utilized to interpolate various models, composing a system to be identified. Differing from the traditional ANFIS, the created technique necessarily represents the system in various well defined regions by unaltered models whose pondered activation as per the membership functions. The selection of regions for the application of the Least Squares method is realized manually from the graphic analysis of the system behavior or from the physical characteristics of the plant. This selection serves as a base to initiate the linear model defining technique and generating the initial configuration of the membership functions. The experiments are conducted in a teaching tank, with multiple sections, designed and created to show the characteristics of the technique. The results from this tank illustrate the performance reached by the technique in task of identifying, utilizing configurations of ANFIS, comparing the developed technique with various models of simple metrics and comparing with the NNARX technique, also adapted to identification

Estrutura ANFIS modificada para identificação e controle de plantas com ampla faixa de operação e não linearidade acentuada

In this work a modification on ANFIS (Adaptive Network Based Fuzzy Inference System) structure is proposed to find a systematic method for nonlinear plants, with large operational range, identification and control, using linear local systems: models and controllers. This method is based on multiple model approach. This way, linear local models are obtained and then those models are combined by the proposed neurofuzzy structure. A metric that allows a satisfactory combination of those models is obtained after the structure training. It results on plant s global identification. A controller is projected for each local model. The global control is obtained by mixing local controllers signals. This is done by the modified ANFIS. The modification on ANFIS architecture allows the two neurofuzzy structures knowledge sharing. So the same metric obtained to combine models can be used to combine controllers. Two cases study are used to validate the new ANFIS structure. The knowledge sharing is evaluated in the second case study. It shows that just one modified ANFIS structure is necessary to combine linear models to identify, a nonlinear plant, and combine linear controllers to control this plant. The proposed method allows the usage of any identification and control techniques for local models and local controllers obtaining. It also reduces the complexity of ANFIS usage for identification and control. This work has prioritized simpler techniques for the identification and control systems to simplify the use of the method

Altered plasma response to zinc and iron tolerance test after Roux-en-Y gastric bypass

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Fatores de influência na produção de leite bovino sobre os níveis da contagem bacteriana total

The objective of this study was to evaluate the influence of milking procedures on the levels of total bacterial count (TBC) in bovine milk. In the first study the influences of procedures for hygienic milking, cleaning of milking equipment and milk cooling tanks on the TBC levels were evaluated. Four bulk samples of milk were collected from each tank in eight properties for TBC analysis, employing the flow cytometry method. A questionnaire was applied in each property to assess the current situation of milking procedures on each production system that took part on this research, followed by training of employees in good agricultural practices in the production of milk and monitoring of the TBC measurements. The methodology for analysis of longitudinal data was considered, focusing on random effects models. The results showed that the handling procedures for milking and the cleanliness of the cooling tank contributed to a further reduction in the levels of TBC raw milk cooling tanks. The second study aimed to describe the percentage of the properties that comply with the Normative Instruction Nº 51 (Brazil s IN 51) with regard to total bacterial count (TBC) in bovine milk. The study was conducted from January 2010 to July 2011. Milk samples were collected from the eight properties selected for TBC analysis by the flow cytometry method. Again, on each property a questionnaire was applied to assess the current situation of milking procedures on each production system that took part on this research, followed by training of employees in good agricultural practices in the production of milk and monitoring of the TBC measurements. The methodology of marginal models based on Generalized Estimate Equations (GEEs) was followed in the statistical analysis. The results showed that the handling procedures of the milking and the cleanliness of the cooling tanks contributed to a considerable percentage of the properties that reached the limits of TBC established by IN 51

The HUman MicroNucleus project on eXfoLiated buccal cells (HUMNXL): The role of life-style, host factors, occupational exposures, health status, and assay protocol

The human buccal micronucleus cytome assay (BMCyt) is one of the most widely used techniques to measure genetic damage in human population studies. Reducing protocol variability, assessing the role of confounders, and estimating a range of reference values are research priorities that will be addressed by the HUMNXL, collaborative study. The HUMNXL, project evaluates the impact of host factors, occupation, life-style, disease status, and protocol features on the occurrence of MN in exfoliated buccal cells. In addition, the study will provide a range of reference values for all cytome endpoints. A database of 5424 subjects with buccal MN values obtained from 30 laboratories worldwide was compiled and analyzed to investigate the influence of several conditions affecting MN frequency. Random effects models were mostly used to investigate MN predictors. The estimated spontaneous MN frequency was 0.74 parts per thousand (95% CI 0.52-1.05). Only staining among technical features influenced MN frequency, with an abnormal increase for non-DNA-specific stains. No effect of gender was evident, while the trend for age was highly significant (p < 0.001). Most occupational exposures and a diagnosis of cancer significantly increased MN and other endpoints frequencies. MN frequency increased in heavy smoking (>= 40 cig/day. FR = 1.37:95% CI 1.03-.82) and decreased with daily fruit consumption (FR = 0.68; 95% CI 0.50-0.91). The results of the HUMNXL, project identified priorities for validation studies, increased the basic knowledge of the assay, and contributed to the creation of a laboratory network which in perspective may allow the evaluation of disease risk associated with MN frequency. (C) 2011 Elsevier B.V. All rights reserved.

Analysis of the electrical conductivity in milking fractions as a mean for detecting and characterizing mastitis in goats

The aim of the work was to study the effect of milking fraction on electrical conductivity of milk (EC) to improve its use in dairy goat mastitis detection using automatic EC measurements during milking. The experiment was carried out on a group of 84 Murciano-Granadina goats (28 primiparous and 56 multiparous). Goats were in the fourth month of lactation. A linear mixed model was used to analyse the relationship between EC or somatic cell count (SCC) of gland milk and parity, mammary gland health status, analysed fraction (first 100 mL=F-1; machine milk=F-2; and stripping milk=F-3) and their first order interactions. Additionally, the mastitis detection characteristics (sensitivity, specificity, positive predictive value and negative predictive value) of SCC and EC were studied at different thresholds.All factors considered were significant for EC and SCC. EC decreased significantly as milking progressed (from F-1 to F-3) in both healthy and infected glands. EC was not significantly different between healthy and infected glands in F-1 and F-2 fractions, but EC of healthy glands (5.01 mS/cm) was significantly lower than in infected glands (5.03 mS/cm) at F-3.Mastitis detection characteristics of EC did not differ amongst studied fractions. The small significant difference of EC between healthy and infected glands obtained in F-3 fraction did not yield better sensitivity results compared to F-1 and F-2. The best EC mastitis detection characteristics were obtained at 5.20 mS/cm threshold (sensitivity of 70% and specificity of 50%). The best SCC mastitis detection characteristics were obtained at 300,000 cells/mL threshold and F-3 fraction (sensitivity of 85% and specificity of 65%).It was concluded that mastitis detection characteristics of EC were similar in the three milking fractions analysed, being slightly better for SCC in F-3 fraction. As shown in previous studies, there are no factors other than the mammary gland health status that affect milk EC and should be considered in the algorithms for mastitis detection to improve the results. (C) 2012 Elsevier B.V. All rights reserved.

Recursive hierarchic segmentation analysis of bone mineral density changes on digital panoramic images

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Componentes de (co)variância e parâmetros genéticos de características de crescimento da raça Simental no Brasil

Informações de genealogia e produção, cedidas pela Associação Brasileira de Criadores da Raça Simental (ABCRS), relativas aos pesos desde o nascimento até um ano de idade, foram utilizadas para estimar, sob modelos alternativos, os componentes de variância e os parâmetros genéticos em animais da raça Simental no Brasil. A matriz de parentesco incluiu 25.812 animais dos quais 7587 com dados de produção. O modelo 1 contém, além do erro, o efeito genético direto. Os modelos seguintes contêm os componentes do modelo 1, mais o efeito permanente de ambiente materno (modelo 2), ou o componente genético materno (modelo 3), ambos os componentes (modelo 5), os componentes do modelo 3 mais a covariância entre os efeitos genéticos direto e materno (modelo 4) e todos os componentes citados (modelo 6). Os modelos foram comparados pelo teste de razão de verossimilhança pelo chi² (P<0,01). Os componentes de variância e os valores de herdabilidades, estimados para os efeitos direto e materno, foram decrescentes, desde o modelo 1 até o modelo 6, na razão direta em que o modelo incorpora mais efeitos aleatórios. Para a fase de aleitamento foi encontrada variância genética nula, entretanto, alto valor para a variância de ambiente permanente. Os efeitos maternos, genético e de ambiente permanente são importantes para a raça Simental no Brasil e devem ser considerados em programas de seleção. Entretanto, os valores mais elevados de herdabilidade materna, encontrados com modelos sem efeito de ambiente permanente, sugerem que o método utilizado não discrimina apropriadamente esses efeitos, oriundos de mesma fonte de variação.

Capacidade materna de cuidar e desnutrição infantil

OBJETIVOS: Identificar e medir a magnitude do risco de desnutrição associada a fatores determinantes da capacidade materna de cuidado infantil: estrutura familiar, escolaridade, trabalho, saúde física e saúde mental maternas. MÉTODOS: Delineou-se um estudo de casos e controles. Foram selecionados 101 casos (crianças com peso/idade abaixo do percentil 5) e 200 controles (crianças com peso/idade acima do percentil 25) mediante inquéritos antropométricos realizados durante três Dias Nacionais de Vacinação, em 1996 e 1997. Os dados foram obtidos em entrevistas realizadas nos domicílios com as mães das crianças. Para detectar o efeito-líquido de cada fator em estudo, realizou-se análise de regressão logística multivariada e hierarquizada. Tais fatores e as possíveis variáveis de controle foram agrupados em blocos, ordenados segundo a precedência com que influiriam sobre o estado nutricional infantil. Adotaram-se p<0,20 para seleção das variáveis de controle (mediante análise univariada) e p<0,05 para identificação de associação estatisticamente significativa entre fatores de estudo e desnutrição infantil. RESULTADOS: Foram identificados como fatores de risco de desnutrição: (a) estrutura familiar adversa indicada pela ausência de companheiro (odds ratio [OR] = 2,2; IC95%, 1,1-4,5); (b) internação materna durante a gravidez (OR=3,5; IC95%, 1,6-7,7); (c) precária saúde mental materna expressa pela presença de três a quatro sintomas de depressão (OR=3,1; IC95%, 0,9-10,3); (d) fatores de estresse familiar, no caso, indícios de alcoolismo em pelo menos um membro da família (OR=2,1; IC95%, 1,2-3,9). A idade da criança no início/retorno da mãe ao trabalho também se associou de modo independente à presença de desnutrição, porém os efeitos variaram: retorno precoce (criança com menos de quatro meses) não significou risco ou proteção; volta da mãe ao trabalho quando a criança tinha entre quatro meses e 12 meses constituiu fator de proteção. CONCLUSÕES: Evidenciou-se que fatores potencialmente definidores da capacidade materna de cuidado exercem efeito independente sobre o estado nutricional infantil.