This study separated two parts: PART ¢¹ Meteorologically adjusted trends of ozone Since meteorological changes strongly affect ambient ozone concentrations, trends in concentrations of ozone upon the adjustment of meteorological variations are important of evaluating emission reduction efforts. This work is to study meteorological effects on the long-term trends of ozone concentration using a multi-variable additive model in Kaohsiung. The long-term trends of ozone concentration were analyzed using the Holland model (without meteorological-adjusted) and the robust MM Regression model (with meteorological-adjusted) based on the data of eight EPA air quality stations from 1997 to 2006 in Kaohsiung area. According to the result of the simulation, the simulated value of the robust MM-Regression model present more valid than the Holland model.The simulated results show that the long-term ozone concentration increases at 13.84% (or 13.06%) monthly (or annually) after meteorological adjustments, less than at 26.10% (or 23.80%) without meteorological adjustments in Kaohsiung county. The simulated results show that the long-term ozone concentration increases at 9.01% (or 6.88%) monthly (or annually) after meteorological adjustments, less than at 22.01% (or 19.67%) without meteorological adjustments in Kaohsiung city. Wind speed, duration of sunshine and pressure are the three dominant factors that influence the ground-level ozone levels in Kaohsiung area. PART ¢º Dispersion of air pollutants in the Hsuehshan Tunnel Concentrations of carbon monoxide (CO) and nitrogen oxides (NOx) were measured from November 14 ¡V 17 2008 in a cross-mountain Hsuehshan traffic tunnel stretching 12.9 km and containing eastward and westward channels. Air pollutants of CO (carbon monoxide) and NOx (nitrogen oxides) will be monitored at the inlet, outlet and vertical shafts of the tunnel. Meanwhile, numerical simulation of three-dimensional turbulent flow will be performed using STAR-CD software. Traffic and pollutant concentrations during the weekends exceeded those during the weekdays. Measured concentrations of CO at the two tunnel outlets (14.5 ¡V 22.8 ppm) were approximately three times higher than those at the two tunnel inlets (3.2 ¡V 7.3 ppm), while concentrations of NOx at the two tunnel outlets (1.9 ¡V 2.9 ppm) were approximately four to five times higher than those at the two tunnel inlets (0.3 ¡V 0.8 ppm). The outlet of vertical draft 2 had the highest pollutant concentrations (CO = 12.3 ppm; NOx = 1.9 ppm), followed by vertical drafts 1 and 3. Three-dimensional turbulence modeling results indicate that airflow in the tunnel was primarily driven by the combined effects of axial fans and vehicles. Results of this study demonstrate that simulated pollutant concentrations increase downstream and are vertically stratified, due to tailpipe exhausts close to tunnel floor. Simulations agreed fairly well with measurements.

In order to personalize or tailor treatments to maximize impact among different subgroups, there is need to model not only the main effects of intervention but also the variation in intervention impact by baseline individual level risk characteristics. To this end a suitable statistical model will allow researchers to answer a major research question: who benefits or is harmed by this intervention program? Commonly in social and psychological research, the baseline risk may be unobservable and have to be estimated from observed indicators that are measured with errors; also it may have nonlinear relationship with the outcome. Most of the existing nonlinear structural equation models (SEM’s) developed to address such problems employ polynomial or fully parametric nonlinear functions to define the structural equations. These methods are limited because they require functional forms to be specified beforehand and even if the models include higher order polynomials there may be problems when the focus of interest relates to the function over its whole domain. To develop a more flexible statistical modeling technique for assessing complex relationships between a proximal/distal outcome and 1) baseline characteristics measured with errors, and 2) baseline-treatment interaction; such that the shapes of these relationships are data driven and there is no need for the shapes to be determined a priori. In the ALV model structure the nonlinear components of the regression equations are represented as generalized additive model (GAM), or generalized additive mixed-effects model (GAMM). Replication study results show that the ALV model estimates of underlying relationships in the data are sufficiently close to the true pattern. The ALV modeling technique allows researchers to assess how an intervention affects individuals differently as a function of baseline risk that is itself measured with error, and uncover complex relationships in the data that might otherwise be missed. Although the ALV approach is computationally intensive, it relieves its users from the need to decide functional forms before the model is run. It can be extended to examine complex nonlinearity between growth factors and distal outcomes in a longitudinal study.

My dissertation consists of three essays in three different areas: international trade; agricultural markets; and nonparametric econometrics. The first and third essays are theoretical papers, while the second essay is empirical. In the first essay, I developed a political economy model of trade agreements where the set of policy instruments are endogenously determined, providing a rationale for countervailing duties (CVDs). Trade-related policy intervention is assumed to be largely shaped in response to rent seeking demand as is often shown empirically. Consequently, the uncertain circumstance during the lifetime of a trade agreement involves both economic and rent seeking conditions. The latter approximates the actual trade policy decisions more closely than the externality hypothesis and thus provides scope for empirical testing. The second essay tests whether normal mixture (NM) generalized autoregressive conditional heteroscedasticity (GARCH) models adequately capture the relevant properties of agricultural commodity prices. Volatility series were constructed for ten agricultural commodity weekly cash prices. NM-GARCH models allow for heterogeneous volatility dynamics among different market regimes. Both in-sample fit and out-of-sample forecasting tests confirm that the two-state NM-GARCH approach performs significantly better than the traditional normal GARCH model. For each commodity, it is found that an expected negative price change corresponds to a higher volatility persistence, while an expected positive price change arises in conjunction with a greater responsiveness of volatility. In the third essay, I propose an estimator for a nonparametric additive unconditional quantile regression model. Unconditional quantile regression is able to assess the possible different impacts of covariates on different unconditional quantiles of a response variable. The proposed estimator does not require d-dimensional nonparametric regression and therefore has no curse of dimensionality. In addition, the estimator has an oracle property in the sense that the asymptotic distribution of each additive component is the same as the case when all other components are known. Both numerical simulations and an empirical application suggest that the new estimator performs much better than alternatives. / the Canadian Agricultural Trade Policy and Competitiveness Research Network, the Structure and Performance of Agriculture and Agri-products Industry Network, and the Institute for the Advanced Study of Food and Agricultural Policy.

The proportional hazards model developed by Cox (1972) is by far the most widely used method for regression analysis of censored survival data. Application of the Cox model to more general event history data has become possible through extensions using counting process theory (e.g., Andersen and Borgan (1985), Therneau and Grambsch (2000)). With its development based entirely on counting processes, Aalen’s additive risk model offers a flexible, nonparametric alternative. Ordinary least squares, weighted least squares and ridge regression have been proposed in the literature as estimation schemes for Aalen’s model (Aalen (1989), Huffer and McKeague (1991), Aalen et al. (2004)). This thesis develops data-driven parameter selection criteria for the weighted least squares and ridge estimators. Using simulated survival data, these new methods are evaluated against existing approaches. A survey of the literature on the additive risk model and a demonstration of its application to real data sets are also provided. / Thesis (Master, Mathematics & Statistics) -- Queen's University, 2007-07-18 22:13:13.243

The main goal of this thesis is to develop Bayesian model for studying the influence of covariate on dependence between random variables. Conditional copula models are flexible tools for modelling complex dependence structures. We construct Bayesian inference for the conditional copula model adapted to regression settings in which the bivariate outcome is continuous or mixed (binary and continuous) and the copula parameter varies with covariate values. The functional relationship between the copula parameter and the covariate is modelled using cubic splines. We also extend our work to additive models which would allow us to handle more than one covariate while keeping the computational burden within reasonable limits. We perform the proposed joint Bayesian inference via adaptive Markov chain Monte Carlo sampling. The deviance information criterion and cross-validated marginal log-likelihood criterion are employed for three model selection problems: 1) choosing the copula family that best fits the data, 2) selecting the calibration function, i.e., checking if parametric form for copula parameter is suitable and 3) determining the number of independent variables in the additive model. The performance of the estimation and model selection techniques are investigated via simulations and demonstrated on two data sets: 1) Matched Multiple Birth and 2) Burn Injury. In which of interest is the influence of gestational age and maternal age on twin birth weights in the former data, whereas in the later data we are interested in investigating how patient’s age affects the severity of burn injury and the probability of death.

The main goal of this thesis is to develop Bayesian model for studying the influence of covariate on dependence between random variables. Conditional copula models are flexible tools for modelling complex dependence structures. We construct Bayesian inference for the conditional copula model adapted to regression settings in which the bivariate outcome is continuous or mixed (binary and continuous) and the copula parameter varies with covariate values. The functional relationship between the copula parameter and the covariate is modelled using cubic splines. We also extend our work to additive models which would allow us to handle more than one covariate while keeping the computational burden within reasonable limits. We perform the proposed joint Bayesian inference via adaptive Markov chain Monte Carlo sampling. The deviance information criterion and cross-validated marginal log-likelihood criterion are employed for three model selection problems: 1) choosing the copula family that best fits the data, 2) selecting the calibration function, i.e., checking if parametric form for copula parameter is suitable and 3) determining the number of independent variables in the additive model. The performance of the estimation and model selection techniques are investigated via simulations and demonstrated on two data sets: 1) Matched Multiple Birth and 2) Burn Injury. In which of interest is the influence of gestational age and maternal age on twin birth weights in the former data, whereas in the later data we are interested in investigating how patient’s age affects the severity of burn injury and the probability of death.

Made available in DSpace on 2016-06-02T18:55:22Z (GMT). No. of bitstreams: 1 4143.pdf: 4028323 bytes, checksum: bb244562f3f5e6280f0390dc28f2aec5 (MD5) Previous issue date: 2011-02-23 / Universidade Federal de Sao Carlos / The Restinga forest is a set of plant communities in mosaic, determined by the characteristics of their substrates as a result of depositional processes and ages. And of all the ecosystems associated with the Atlantic, the most fragile and susceptible to human disturbance. In this complex mosaic are the physiognomies of restinga forests of high-stage regeneration (high restinga) and middle stage of regeneration (low restinga), each with its plant characteristics that differentiate them. Located on the coastal plains of the Brazilian coast, suffering internal influences both the continental slopes, as well as navy. His solo comes from the Quaternary and are subject to constant deposition of sediments. The climate on the coast, according to Köppen classification, type is tropical. In recent decades, with growing concern about natural resources and environmental quality, intensified research, resulting in the definition of soil quality (SQ), strongly rooted in the concept of sustainability. To this end, several models have been proposed in an attempt to assess soil quality index (SQI). The monitoring of soil quality should be directed to detect trends that are measurable changes in a period of time. The objectives of this study were: a) Comparative evaluation of the characterization of soil fertility, through chemical and physical parameters under restinga forest of high and low, with respect to distribution of the root in the soil profile, and b) Determine the index Soil Quality for restinga forest in high-and middle-stage regeneration and a resting area with no vegetation. This work was conducted in four locations: (1) Anchieta Island, Ubatuba, (2) Juréia-Itatins Ecological Station, Iguape, (3) Vila das Pedrinhas, Comprida Island; and (4) Cardoso Island, Cananeia. Studies on soil fertility have been made in depths of 0 to 5, 00-10, 00-20, 20-40 and 40 to 60cm for the chemical and physical analysis, with five replicates for each vegetation type, for each study site, each composed of twelve subsamples. Also being evaluated the distribution of the root in the soil profile. To determine the rate of soil quality, chemical analysis were made, microbiological and physical-layer 0-10cm depth. Using two models in determining the rate of soil quality: Additive Model (MA) and Comparative Additive Model (MAC). It is concluded that the root system for all studied vegetation types found in the more superficial layers, 0-10 and 10-20cm, mainly in the 0-10cm (80%), that low levels of calcium and elevated aluminum restrict root development. All the studied have low soil fertility, with base saturation values below 16%, where most of these environments CEC is occupied by aluminum. The additive model produces quantitative results and the additive model comparative quantitative and qualitative results (ground potential), the SQI values were obtained by the MAC for all local and low vegetation types and realistic, demonstrating the low potential for biomass production in these soils, and its low resilience. Values similar to the forests with and without vegetation showed numerically the consideration that the restinga is an edaphic vegetation. And that the use of routine chemical analysis is sufficient to determine the IQS. / A Restinga é um conjunto de comunidades vegetais em mosaico, determinadas pelas características de seus substratos resultantes de processos deposicionais e idades. De todos os ecossistemas associados à Mata Atlântica, a restinga é a mais frágil e susceptível às perturbações antrópicas. Neste complexo mosaico encontram-se as fitofisionomias de florestas de restinga em estágio de elevada regeneração (restinga alta) e em estágio de média regeneração (restinga baixa), cada qual com suas características vegetais que as diferenciam. Localizam-se nas planícies costeiras do litoral brasileiro, sofrendo influência tanto das encostas internas continentais, bem como marinha. Seu solo tem origem no quaternário e estão sujeitos a constantes deposição de sedimentos. O clima no litoral, segundo classificação de Köppen, é do tipo tropical. Nas últimas décadas, com a crescente preocupação com os recursos naturais e a qualidade do meio ambiente, intensificaram-se as pesquisas, resultando na definição do conceito de Qualidade do Solo (QS), fortemente alicerçado no conceito de sustentabilidade. Para tanto, vários modelos foram propostos na tentativa de avaliar um Índice de Qualidade do Solo (IQS). O monitoramento da qualidade do solo deve ser orientado para detectar tendências de mudanças que são mensuráveis num período de tempo. Os Objetivos deste estudo foram: a) avaliação comparativa da caracterização da fertilidade do solo, através dos parâmetros químicos e físicos, sob floresta de restinga alta e baixa, com relação a distribuição do sistema radicular no perfil do solo; e b) determinar o Índice de Qualidade do Solo para floresta de restinga em estágio de elevada e média regeneração e para uma área de restinga sem vegetação. Este estudo foi realizado em quatro locais: (1) Parque Estadual da Ilha Anchieta, município de Ubatuba; (2) Estação Ecológica Juréia-Itatins, Estação Ecológica dos Chauás, município de Iguape; (3) Vila de Pedrinhas no município de Ilha Comprida; e (4) Parque Estadual da Ilha do Cardoso, município de Cananéia. Os estudos sobre fertilidade do solo foram feitos nas profundidades de 0 a 5, 0 a 10, 0 a 20, 20 a 40 e 40 a 60cm para as análises químicas e físicas, com cinco repetições para cada fitofisionomia, para cada local de estudo, cada uma delas composta por doze subamostras. Também foi avaliada a distribuição do sistema radicular no perfil do solo. Para a determinação do índice de qualidade do solo (IQS), foram feitas analises químicas, físicas e microbiológicas na camada de 0-10cm de profundidade. Utilizaram-se dois modelos na determinação do índice de qualidade do solo: Modelo Aditivo (MA) e Modelo Aditivo Comparativo (MAC). Conclui-se que o sistema radicular para todas as fitofisionomias estudadas encontra-se nas camadas mais superficiais, 0-10 e 10-20cm, principalmente na camada de 0-10cm (80%), e que os teores baixos de cálcio e elevados de alumínio restringem o desenvolvimento radicular. Todos os ambientes estudados apresentaram baixa fertilidade do solo, com valores de saturação por bases inferiores a 18%, onde a maior parte da CTC destes ambientes está ocupada por alumínio. O modelo aditivo produz resultados quantitativos e o modelo aditivo comparativo resultados quantitativos e qualitativos (potencial do solo), que os valores de IQS obtidos pelo MAC foram baixos e realísticos para todos os locais e fitofisionomias, demonstrando o baixo potencial de produção de biomassa desses solos, bem como sua baixa resiliência. Os valores semelhantes para as florestas com e sem vegetação demonstraram que a restinga é uma vegetação edáfica, e que o emprego de análise química de rotina é suficiente na determinação do índice de qualidade do solo.

Submitted by Irene Nascimento (irene.kessia@ufpe.br) on 2016-06-22T16:10:58Z No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Dissertacao_SinndyDayanaRicoLugo_2016.pdf: 3067162 bytes, checksum: bf4e6bc8fdefc54a21c8315582de00a1 (MD5) / Made available in DSpace on 2016-06-22T16:10:58Z (GMT). No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Dissertacao_SinndyDayanaRicoLugo_2016.pdf: 3067162 bytes, checksum: bf4e6bc8fdefc54a21c8315582de00a1 (MD5) Previous issue date: 2016-02-01 / CNPQ / A determinação dos requerimentos de materiais nas Pequenas e Médias Empresas (PMEs) cujo ambiente de produção é do tipo Job Shop, tem sido categorizado na literatura como um problema devido ao complexo processo de tomada de decisões subjacente, gerado pela grande quantidade de variáveis no sistema de fabricação, aos níveis de apropriação de tecnologias da informação e às características dos modelos e das ferramentas que atualmente encontram-se disponíveis. Dentro deste contexto, uma solução fundamentada em um modelo de decisão multicritério foi proposta, incluindo a execução do processo de elicitação das preferências do decisor, e suportada na geração de um Sistema de Apoio à Decisão (SAD) de ambiente Web. Ao longo deste trabalho é apresentada a caracterização das etapas de construção do modelo, os pontos relevantes para a escolha do Modelo Aditivo como base, as melhorias feitas ao processo de elicitação, e o detalhamento da interação do software desenvolvido com o processo decisório de determinação de requerimentos de matérias primas. Apresenta-se também a aplicação do modelo em algumas empresas do tipo PME, realizando uma análise comparativa entre os resultados esperados e os obtidos com o uso da ferramenta SAD e recolhendo todos os comentários dos decisores, com a finalidade de caracterizar, em um ambiente de fábrica real, os prós, contras e possíveis melhorias do modelo proposto. Todas as aplicações foram realizadas em duas fases: na primeira o decisor usou o SAD de forma isolada, sem o acompanhamento da analista com o intuito de obter uma visão totalmente externa; e na segunda o decisor usou o software com o acompanhamento direto da analista tendo como objetivo a interatividade e a troca de conhecimentos. A execução da primeira fase proporcionou informação relevante de como os decisores se sentem em relação às perguntas da elicitação de preferências, à linguagem usada, aos gráficos e às demais características desenvolvidas no aplicativo, concluindo que não contar com um analista obriga ao decisor a pensar cuidadosamente nas suas respostas e a ler detalhadamente as instruções. A segunda fase permitiu aos decisores maior compreensão do processo de elicitação e principalmente, em relação ao uso do SAD na etapa da analise de sensibilidade. Adicionalmente, apresenta-se a proposta de um segundo modelo baseado em outras teorias de decisão multicritério, Teoria de Utilidade Multiatributo (MAUT por sua sigla em inglês) e Utilidade Rank Dependente (RDU por sua sigla em inglês), com a diferença de que a ferramenta SAD foi testada com dados reais de única empresa. Assim, os resultados da aplicação deste modelo mostram diferenças substanciais entre utilizar o método clássico da Utilidade Esperada (EU em inglês) e usar a RDU; enquanto que diferenças menores, mas também relevantes, foram encontradas entre elicitar a função peso da probabilidade e usar os valores dos parâmetros sugeridos comumente na literatura com base em estudos comportamentais / Determining the requirements of materials in Small and Medium Enterprises (SMEs), whose production environment is Job Shop type, has been categorized in the literature as a problem caused by the complex underlying decision-making process, generated by the large number of variables in the manufacturing system, the appropriation levels of information technology and the characteristics of the models and tools that currently are available. Within this framework, a solution based on a multi-criteria decision model was proposed, including the execution of the elicitation process of decision maker's preferences, and supported in the generation of a decision support system (DSS) Web based. Throughout this document presents the characterization of model building steps, relevant points for choosing the Additive Model as a base, improvements made to the elicitation process and details of interaction within the software developed and the decision-making process of raw material requirements determination. It presents also the application of the model in some companies of type SME, performing a comparative analysis between expected results and those obtained using the SAD tool and compiling all the comments of decision-makers, in order to characterize, in environment real factory, the pros, cons and possible improvements of the proposed model. All applications were done in two phases: first the decision maker used the SAD in isolation, without the accompaniment of the analyst in order to get a fully external view; and in the second the decision maker used the software with analyst's direct monitoring aiming to the interaction and exchange of knowledge. The implementation of the first phase provided relevant information about how the decision-makers feel in relation to the preferences elicitation questions, to the language used, to the graphics and to other features developed in the application, concluding that if there is not an analyst, the decision maker has to think carefully in their responses and thoroughly reads the instructions. The second phase offered to decision-makers greater understanding of the elicitation process and especially regarding the use of the SAD in the sensitivity analysis step. In addition, a second model based on other theories of multi-criteria decision (Multi-attribute Utility Theory (MAUT) and Rank Dependent Utility (RDU)) was presented, with the difference that the SAD tool was tested with real data of only one company. Thus, the results of applying this model show substantial differences between using the classic method of Expected Utility (EU) and use RDU; while minor differences, but also relevant, were found between eliciting the probability weighting function and using the values of the parameters commonly suggested in the literature based on behavioral studies.

Orientador: Fernando Roberto Martins / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Biologia / Made available in DSpace on 2018-08-23T02:32:50Z (GMT). No. of bitstreams: 1 VanMelis_Juliano_D.pdf: 2552550 bytes, checksum: 8227a941fa221a10cce8b272ae92449f (MD5) Previous issue date: 2013 / Resumo: Apesar da importância que as trepadeiras apresentam em florestas tropicais, estudos sobre a montagem da comunidade de lianas (trepadeiras lenhosas e sublenhosas) que investiguem desde a contribuição dos fatores abióticos e bióticos até fatores intrínsecos (coexistência entre indivíduos) são escassos. O objetivo geral desta tese é pesquisar a estruturação da comunidade das espécies de lianas em uma Floresta Estacional Semidecídua (FES), investigando (1) a importância relativa dos fatores ambientais e espaciais para diferentes espécies de lianas, (2) a estruturação filogenética da comunidade de trepadeiras em diferentes ambientes, e (3) os efeitos diretos ou mediados das árvores e arbustos para o número de espécies e indivíduos de trepadeiras. Mostramos que (1) grande parte da variação na composição de espécies de lianas em uma FES é devido a fatores não investigados (fatores estocásticos) e o espaço (autocorrelação espacial). Portanto, concluímos que os maiores determinantes na variação da composição de espécies de lianas em uma FES é a aleatoriedade (sendo reflexo da variação estocástica das populações) e a limitação por dispersão (demonstrada pela alta autocorrelação espacial). No segundo capítulo (2), encontramos que uma maioria discreta das parcelas apresentou maior aproximação filogenética do que o esperado ao acaso na comunidade de trepadeiras amostrada. Houve pouca influência de variáveis relacionadas à dinâmica florestal na variação da aproximação filogenética, sendo que áreas com árvores mais altas e maior proporção de árvores do presente apresentavam maior aproximação filogenética que outras áreas. Concluímos que em áreas de dossel mais baixo e menor proporção de árvores do presente (clareiras) não apresentam menor sinal filogenético, pois todas as espécies de lianas apresentariam potencial de existirem nestas áreas, enquanto que nas áreas de floresta madura haveria a existência de filtros ambientais para a existência de poucos ramos filogenéticos. Por último (3), encontramos que os atributos da comunidade de árvores e arbustos são fatores importantes na variação dos atributos da comunidade de lianas, sendo parte dele decorrente do distúrbio no dossel. Mas o distúrbio no dossel como fator direto é mais importante na variação da abundância e número de espécies de lianas em uma Floresta Estacional Semidecídua / Abstract: Despite the fact that climbing plants present in tropical forests, studies which investigate the contribution of abiotic and biotic factors or intrinsic factors (coexistence between individuals) on community assembly of lianas (woody and sub-woody climbers) are scarce. The overall objective of this thesis is to research the community structure of liana species in a Seasonal Semideciduous Forest (SSF), investigating (1) the relative importance of environmental and spatial factors on community assembly of lianas, (2) the phylogenetic structure of climbing plants community along the forest development (treefall gaps to old-growth forest), and (3) the direct or indirect effects of trees and shrubs for the number of species and individuals of climbing plants. We show that (1) much of the variation in species composition of lianas in a SSF is due to stochastic factors and space. Therefore, we conclude that the major determinants of variation in lianas' species composition in a TSF are stochastic variance of populations, shown by the unexplained factors, and dispersion limitation, shown by spatial autocorrelation. In the second chapter (2), we found that a slight majority of the sample plots showed cluster phylogenetic structure in the climbing plants community. There was a slight influence of variables related to forest dynamics in the variation of the phylogenetic structure, and areas with tall trees and higher proportion of present trees had higher values of clustering in phylogenetic structure than other areas. We conclude that in areas of lower canopy and smaller proportion of present trees (treefall gaps) showed few phylogenetic branches, since all species of climbing plants would be existing in these areas, while areas of old-growth forest would demonstrate environmental filters for the climbing plants. Finally, we also found (3) that the community of trees and shrubs' attributes (abundance and species richness) are important factors in the variation of attributes liana community (species richness and abundance), being part of it due to the canopy disturbance. But canopy disturbance was the more important direct factor in variance of abundance and species richness of lianas in a Seasonal Semideciduous Forest / Doutorado / Doutor em Biologia Vegetal

L'électricité ne se stockant pas aisément, EDF a besoin d'outils de prévision de consommation et de production efficaces. Le développement de nouvelles méthodes automatiques de sélection et d'estimation de modèles de prévision est nécessaire. En effet, grâce au développement de nouvelles technologies, EDF peut étudier les mailles locales du réseau électrique, ce qui amène à un nombre important de séries chronologiques à étudier. De plus, avec les changements d'habitude de consommation et la crise économique, la consommation électrique en France évolue. Pour cette prévision, nous adoptons ici une méthode semi-paramétrique à base de modèles additifs. L'objectif de ce travail est de présenter des procédures automatiques de sélection et d'estimation de composantes d'un modèle additif avec des estimateurs en plusieurs étapes. Nous utilisons du Group LASSO, qui est, sous certaines conditions, consistant en sélection, et des P-Splines, qui sont consistantes en estimation. Nos résultats théoriques de consistance en sélection et en estimation sont obtenus sans nécessiter l'hypothèse classique que les normes des composantes non nulles du modèle additif soient bornées par une constante non nulle. En effet, nous autorisons cette norme à pouvoir converger vers 0 à une certaine vitesse. Les procédures sont illustrées sur des applications pratiques de prévision de consommation électrique nationale et locale.Mots-clés: Group LASSO, Estimateurs en plusieurs étapes, Modèle Additif, Prévision de charge électrique, P-Splines, Sélection de variables / French electricity load forecasting encounters major changes since the past decade. These changes are, among others things, due to the opening of electricity market (and economical crisis), which asks development of new automatic time adaptive prediction methods. The advent of innovating technologies also needs the development of some automatic methods, because we have to study thousands or tens of thousands time series. We adopt for time prediction a semi-parametric approach based on additive models. We present an automatic procedure for covariate selection in a additive model. We combine Group LASSO, which is selection consistent, with P-Splines, which are estimation consistent. Our estimation and model selection results are valid without assuming that the norm of each of the true non-zero components is bounded away from zero and need only that the norms of non-zero components converge to zero at a certain rate. Real applications on local and agregate load forecasting are provided.Keywords: Additive Model, Group LASSO, Load Forecasting, Multi-stage estimator, P-Splines, Variables selection