Analyses of data with missing values often require assumptions about missingness mechanisms that cannot be assessed empirically, highlighting the need for sensitivity analyses. However, universal recommendations for reporting missing data and conducting sensitivity analyses in empirical studies are scarce. Both steps are often neglected by practitioners due to the lack of clear guidelines for summarizing missing data and systematic explorations of alternative assumptions, as well as the typical attendant complexity of missing not at random (MNAR) models. We propose graphical displays that help visualize and systematize the results of sensitivity analyses, building upon the idea of "tipping-point" analysis for experiments with dichotomous treatment. The resulting "enhanced tipping-point displays" (ETP) are convenient summaries of conclusions drawn from using different modeling assumptions about the missingness mechanisms, applicable to a broad range of outcome distributions. We also describe a systematic way of exploring MNAR models using ETP displays, based on a pattern-mixture factorization of the outcome distribution, and present a set of sensitivity parameters that arises naturally from such a factorization. The primary goal of the displays is to make formal sensitivity analyses more comprehensible to practitioners, thereby helping them assess the robustness of experiments' conclusions. We also present an example of a recent use of ETP displays in a medical device clinical trial, which helped lead to FDA approval. The last part of the dissertation demonstrates another method of sensitivity analysis in the same clinical trial. The trial is complicated by missingness in outcomes "due to death", and we address this issue by employing Rubin Causal Model and principal stratification. We propose an improved method to estimate the joint posterior distribution of estimands of interest using a Hamiltonian Monte Carlo algorithm and demonstrate its superiority for this problem to the standard Metropolis-Hastings algorithm. The proposed methods of sensitivity analyses provide new collections of useful tools for the analysis of data sets plagued with missing values. / Statistics

Les estimateurs pondérés généralement utilisés pour analyser les enquêtes cas-cohorte ne sont pas pleinement efficaces. Or, les enquêtes cas-cohorte sont un cas particulier de données incomplètes où le processus d'observation est contrôlé par les organisateurs de l'étude. Ainsi, des méthodes d'analyse pour données manquant au hasard (MA) peuvent être pertinentes, en particulier, l'imputation multiple, qui utilise toute l'information disponible et permet d'approcher l'estimateur du maximum de vraisemblance partielle.Cette méthode est fondée sur la génération de plusieurs jeux plausibles de données complétées prenant en compte les différents niveaux d'incertitude sur les données manquantes. Elle permet d'adapter facilement n'importe quel outil statistique disponible pour les données de cohorte, par exemple, l'estimation de la capacité prédictive d'un modèle ou d'une variable additionnelle qui pose des problèmes spécifiques dans les enquêtes cas-cohorte. Nous avons montré que le modèle d'imputation doit être estimé à partir de tous les sujets complètement observés (cas et non-cas) en incluant l'indicatrice de statut parmi les variables explicatives. Nous avons validé cette approche à l'aide de plusieurs séries de simulations: 1) données complètement simulées, où nous connaissions les vraies valeurs des paramètres, 2) enquêtes cas-cohorte simulées à partir de la cohorte PRIME, où nous ne disposions pas d'une variable de phase-1 (observée sur tous les sujets) fortement prédictive de la variable de phase-2 (incomplètement observée), 3) enquêtes cas-cohorte simulées à partir de la cohorte NWTS, où une variable de phase-1 fortement prédictive de la variable de phase-2 était disponible. Ces simulations ont montré que l'imputation multiple fournissait généralement des estimateurs sans biais des risques relatifs. Pour les variables de phase-1, ils approchaient la précision obtenue par l'analyse de la cohorte complète, ils étaient légèrement plus précis que l'estimateur calibré de Breslow et coll. et surtout que les estimateurs pondérés classiques. Pour les variables de phase-2, l'estimateur de l'imputation multiple était généralement sans biais et d'une précision supérieure à celle des estimateurs pondérés classiques et analogue à celle de l'estimateur calibré. Les résultats des simulations réalisées à partir des données de la cohorte NWTS étaient cependant moins bons pour les effets impliquant la variable de phase-2 : les estimateurs de l'imputation multiple étaient légèrement biaisés et moins précis que les estimateurs pondérés. Cela s'explique par la présence de termes d'interaction impliquant la variable de phase-2 dans le modèle d'analyse, d'où la nécessité d'estimer des modèles d'imputation spécifiques à différentes strates de la cohorte incluant parfois trop peu de cas pour que les conditions asymptotiques soient réunies.Nous recommandons d'utiliser l'imputation multiple pour obtenir des estimations plus précises des risques relatifs, tout en s'assurant qu'elles sont analogues à celles fournies par les analyses pondérées. Nos simulations ont également montré que l'imputation multiple fournissait des estimations de la valeur prédictive d'un modèle (C de Harrell) ou d'une variable additionnelle (différence des indices C, NRI ou IDI) analogues à celles fournies par la cohorte complète

The release of the Illumina BovineSNP50 BeadChip in late 2007 has drawn attention from cattle breeders around the world to develop breeding programs that leverage association of these single nucleotide polymorphism (SNP) with economically important quantitative trait loci (QTL). In that context this project has come to study applications of the SNP panel in beef cattle. Analysis of linkage disequilibrium (LD) existing in Angus, Charolais, and crossbred animals revealed the pattern of LD within each breed group, as well as the persistence of LD phase between pairs of the breed groups. This is important for genomic selection where SNP are trained in one population and used to predict breeding value for animals in another population. Detection of chromosome regions potentially carrying QTL or causative mutations affecting the phenotypic variation in economically important traits was presented at individual SNP and haplotype levels. There were 269 SNP associated (P<0.001) with birth weight (BWT), weaning weight (WWT), average daily gain (ADG), dry matter intake (DMI), mid-test metabolic weight (MMWT), residual feed intake (RFI). They explained 1.64% - 8.06% of the phenotypic variation in these traits. There were 520 SNP associated (P<0.001) with carcass quality traits, namely hot carcass weight, back fat thickness, ribeye area, marbling scores, lean yield grade by Beef Improvement Federation, steak tenderness, and six rib dissection traits. These SNP explained 1.90 - 5.89% of the phenotypic variance of the traits. Many of the significant SNP were located on chromosome 6. Six haplotypes were found associated (P<0.05) with ADG, DMI, and RFI. In order for genomic selection to happen in beef cattle, higher density SNP panels should be made available at low genotyping cost. However, the cost of genotyping animals for high density SNP chip is still high, thus genotype imputation has come to practice. The last chapter of this thesis compared two approaches presently used in genotype imputation, investigated factors affecting imputation accuracy, as well as the impact of imputation accuracy on genomic estimated breeding value (GEBV). It proved that the highest possible accuracy of GEBV is attainable with sufficiently large groups of reference animals. / Ontario Ministry of Agriculture, Food and Rural Affairs. Ontario Cattlemen’s Association. Ontario Farm Innovation Program. Agriculture and Agri-Food Canada’s Growing Forward Program. Agriculture Adaptation Council. Ontario Research and Development Program. MITACS Accelerate. Beef Improvement Opportunities.

Made available in DSpace on 2015-05-14T12:47:12Z (GMT). No. of bitstreams: 1 arquivototal.pdf: 2080872 bytes, checksum: 97ca6e77578ad42b13570272cbc34e7e (MD5) Previous issue date: 2012-02-24 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / The quality of information has been of particular interest in health. It is known that the incompleteness of information is a very common problem in information systems and epidemiological studies. Thus, it has been imputation as a solution of the missing data, which are created artificially complete set of data subject to statistical analysis. This study aimed to analyze the data quality HIPERDIA items on satisfaction and hypertensive patients of the Family Health Strategy in the city of João Pessoa / PB on the service and the use of imputation methods for missing data. Secondary data were obtained from duplicate HIPERDIA, the hypertensive patients enrolled between 2006/2007 in 36 family health teams, resulting in a representative sample of 343 users in the city of João Pessoa / PB. As a primary source was constructed an instrument consisting of eight core dimensions of primary care, measured on Likert scale ranging from "0" to "5". The techniques were applied to the method of Single Imputation: Replacement for Central Value Trend (TC), Hot Deck, Estimated Maximum Likelihood (MV) and Multinomial Logistic Regression (RLM), were compared using the percentage of correct answers, average error square (RMSE) and mean absolute percentage error (MAPE). Was built to simulate two different scenarios sample with different proportions of missing data (5%, 10%, 15%, 30% and 40%). The comparison of the allocation methods, for variable setting with a type having overlapping response to the other, the method was that TC gave better performance, followed by the method of RLM. For the scenario with homogeneous frequency response, the best method was to RLM. The study has demonstrated that there are still errors in the completion of HIPERDIA and that allowed us to recover the imputation characteristics of the representation of the original data, verifying that the imputation methods adopted brought reliability and reduction of bias in the sample proportions of up to 40% of missing data. / A qualidade das informações tem sido objeto de interesse particularmente na área da saúde. Sabe-se que a incompletude de informações é um problema muito comum nos sistemas de informação e em estudos epidemiológicos. Desta forma, tem-se como solução a imputação de dados, onde são criados conjunto de dados artificialmente completos passíveis de análise estatística. Esse estudo objetivou analisar a qualidade dos dados do HIPERDIA e dos itens sobre satisfação de usuários hipertensos da Estratégia Saúde da Família no município de João Pessoa/PB sobre o serviço e o uso de métodos de imputação para dados faltantes. Os dados secundários foram obtidos da segunda via do HIPERDIA, dos hipertensos cadastrados entre 2006/2007 em 36 equipes de Saúde da Família, resultando numa amostra representativa de 343 usuários do município de João Pessoa/PB. Como fonte primária foi construído um instrumento composto por 8 dimensões essenciais da atenção básica, mensurados na Escala de Likert variando de 0 a 5 . As técnicas foram aplicadas para o método de Imputação Única: Substituição por um Valor de Tendência Central (TC), Hot Deck, Estimativa de Máxima Verossimilhança (MV) e Regressão Logística Multinomial (RLM), sendo comparados através do percentual de acerto, erro médio quadrado (RMSE) e erro percentual médio absoluto (MAPE). Foi construída a simulação de dois cenários amostrais distintos com diferentes proporções de dados faltantes (5%,10%, 15%, 30% e 40%). Na comparação dos métodos de imputação, para cenário com variável apresentando um tipo de resposta sobrepondo às outras, o método de TC foi o que obteve melhor performance, seguido do método de RLM. Para o cenário com homogeneidade de frequencia de respostas, o melhor método foi o de RLM. O estudo permitiu demonstrar que ainda existem falhas no preenchimento do HIPERDIA e que a imputação permitiu resgatar as características da representação dos dados originais, verificando que os métodos de imputação adotados trouxeram fidedignidade e diminuição de vieses na amostra para proporções de até 40% de dados faltantes.

Objetivou-se com este trabalho: (1) avaliar parâmetros de diversidade genética e de estrutura populacional com base nas raças Angus, Devon, Hereford e Shorthorn. Taxa de endogamia (ΔF), tamanho efetivo (Ne), grau de parentesco, entre outros parâmetros, foram estimados para fornecer subsídios aos programas de melhoramento. Os parâmetros indicaram adequada diversidade genética, com Ne variando entre 128 no Devon e 303 no Shorthorn e ΔF variando entre 1,50 no Angus e 3,92 no Devon; (2) avaliar estratégias de imputação de genótipos utilizando dados de Braford e Hereford através de painéis de baixa densidade (3K, 6K, 8K, 15K e 20K) para os painéis de 50K e 777K. Para o painel de 777K, também foram utilizados na imputação os painéis de 50K, 90iK e 90tK. Os resultados indicaram que, com exceção do painel de 3K, todos os demais painéis de baixa densidade poderiam ser utilizados como base visando à imputação para o painel de 50K e também que os painéis de média densidade (50K, 90iK e 90tK), poderiam ser utilizados como base na imputação para o painel de 777K. Esses painéis mostraram-se eficientes e possuem, em geral, custos compatíveis com a atividade pecuária; (3) avaliar a acurácia de predição dos valores genômicos utilizando alguns painéis de baixa densidade (8K e 15K) imputados para o painel de 50K, relacionando os resultados com o uso do painel original de 50K. A acurácia do valor genômico direto (DGV) e do valor genético genômico (GEBV) com o valor genético (EBV) utilizando painéis imputados ou não, indicaram que não houveram diferenças em acurácia e as perdas em acurácia por utilizar os painéis imputados ficaram entre -0,0002 e -0,0021 dependendo do painel, do cenário e da característica analisada; (4) usar marcadores moleculares na seleção genômica testando dois métodos BLUP (procedimento de passo único ou de multi passo) com dados simulados de bovinos de corte. Os resultados demonstraram, com base nos parâmetros estudados, igualdade de resultados entre os dois procedimentos; (5) avaliar a viabilidade do uso da seleção genômica usando dados de campo de animais Braford e Hereford, testando os dois métodos BLUP (passo único e multi passo). Os resultados com base nos parâmetros estudados, mostraram que as acurácias de predição do DGV e do GEBV foram iguais nos dois procedimentos, porém no método multi passo as predições genômicas foram menos viesadas. / The aim of this work were: (1) to evaluate parameters of genetic diversity and population structure based on Angus, Devon, Hereford and Shorthorn breeds. Inbreeding rate (ΔF), effective size (Ne), relatedness, among other parameters, were estimated to provide subsidies for breeding programs. The parameters indicated a good genetic diversity, with Ne ranging from 128 in Devon to 303 in Shorthorn and ΔF ranging from 1.50 in Angus to 3.92 in Devon; (2) to evaluate strategies of genotype imputation with Braford and Hereford beef data using low density panels (3K, 6K, 8K, 15K and 20K) for 50K and 777K panels. For imputation to the 777K panel were also used the 50K, 90iK and 90tK panels. The results indicated that, except for the 3K panel, all other low density panels could be used of imputation the 50K panel and also the medium density panels (50K, 90iK and 90tK), could be used of imputation the 777K panel. These panels have been efficient and have, in general, compatible costs of the beef cattle operation; (3) to evaluate the accuracy of genomic prediction using some low density panels (8K and 15K) imputed to the 50K panel, relating the results with 50K original panel. The accuracy of direct genomic value (DGV) and genomic estimated breeding value (GEBV) with estimated breeding value (EBV) using imputed or not panels, indicated that there were no differences in accuracy and the losses in accuracy by using the imputed panels ranged from -0.0002 to -0.0021 depending on the panel, the scenario and the trait; (4) to use molecular markers in genomic selection testing two BLUP methods (single and two steps) with simulated beef cattle data. The results showed, based on the parameters studied, equality of results between the two methods; (5) to evaluate the viability of using the genomic selection using Braford and Hereford beef cattle and testing the two BLUP methods (single and two steps). The results, based on the parameters studied, showed that DGV and GEBV accuracies were similar in both methods, but the genomic predictions were less biased with then two-step method.

Objetivou-se com este trabalho: (1) avaliar parâmetros de diversidade genética e de estrutura populacional com base nas raças Angus, Devon, Hereford e Shorthorn. Taxa de endogamia (ΔF), tamanho efetivo (Ne), grau de parentesco, entre outros parâmetros, foram estimados para fornecer subsídios aos programas de melhoramento. Os parâmetros indicaram adequada diversidade genética, com Ne variando entre 128 no Devon e 303 no Shorthorn e ΔF variando entre 1,50 no Angus e 3,92 no Devon; (2) avaliar estratégias de imputação de genótipos utilizando dados de Braford e Hereford através de painéis de baixa densidade (3K, 6K, 8K, 15K e 20K) para os painéis de 50K e 777K. Para o painel de 777K, também foram utilizados na imputação os painéis de 50K, 90iK e 90tK. Os resultados indicaram que, com exceção do painel de 3K, todos os demais painéis de baixa densidade poderiam ser utilizados como base visando à imputação para o painel de 50K e também que os painéis de média densidade (50K, 90iK e 90tK), poderiam ser utilizados como base na imputação para o painel de 777K. Esses painéis mostraram-se eficientes e possuem, em geral, custos compatíveis com a atividade pecuária; (3) avaliar a acurácia de predição dos valores genômicos utilizando alguns painéis de baixa densidade (8K e 15K) imputados para o painel de 50K, relacionando os resultados com o uso do painel original de 50K. A acurácia do valor genômico direto (DGV) e do valor genético genômico (GEBV) com o valor genético (EBV) utilizando painéis imputados ou não, indicaram que não houveram diferenças em acurácia e as perdas em acurácia por utilizar os painéis imputados ficaram entre -0,0002 e -0,0021 dependendo do painel, do cenário e da característica analisada; (4) usar marcadores moleculares na seleção genômica testando dois métodos BLUP (procedimento de passo único ou de multi passo) com dados simulados de bovinos de corte. Os resultados demonstraram, com base nos parâmetros estudados, igualdade de resultados entre os dois procedimentos; (5) avaliar a viabilidade do uso da seleção genômica usando dados de campo de animais Braford e Hereford, testando os dois métodos BLUP (passo único e multi passo). Os resultados com base nos parâmetros estudados, mostraram que as acurácias de predição do DGV e do GEBV foram iguais nos dois procedimentos, porém no método multi passo as predições genômicas foram menos viesadas. / The aim of this work were: (1) to evaluate parameters of genetic diversity and population structure based on Angus, Devon, Hereford and Shorthorn breeds. Inbreeding rate (ΔF), effective size (Ne), relatedness, among other parameters, were estimated to provide subsidies for breeding programs. The parameters indicated a good genetic diversity, with Ne ranging from 128 in Devon to 303 in Shorthorn and ΔF ranging from 1.50 in Angus to 3.92 in Devon; (2) to evaluate strategies of genotype imputation with Braford and Hereford beef data using low density panels (3K, 6K, 8K, 15K and 20K) for 50K and 777K panels. For imputation to the 777K panel were also used the 50K, 90iK and 90tK panels. The results indicated that, except for the 3K panel, all other low density panels could be used of imputation the 50K panel and also the medium density panels (50K, 90iK and 90tK), could be used of imputation the 777K panel. These panels have been efficient and have, in general, compatible costs of the beef cattle operation; (3) to evaluate the accuracy of genomic prediction using some low density panels (8K and 15K) imputed to the 50K panel, relating the results with 50K original panel. The accuracy of direct genomic value (DGV) and genomic estimated breeding value (GEBV) with estimated breeding value (EBV) using imputed or not panels, indicated that there were no differences in accuracy and the losses in accuracy by using the imputed panels ranged from -0.0002 to -0.0021 depending on the panel, the scenario and the trait; (4) to use molecular markers in genomic selection testing two BLUP methods (single and two steps) with simulated beef cattle data. The results showed, based on the parameters studied, equality of results between the two methods; (5) to evaluate the viability of using the genomic selection using Braford and Hereford beef cattle and testing the two BLUP methods (single and two steps). The results, based on the parameters studied, showed that DGV and GEBV accuracies were similar in both methods, but the genomic predictions were less biased with then two-step method.

Objetivou-se com este trabalho: (1) avaliar parâmetros de diversidade genética e de estrutura populacional com base nas raças Angus, Devon, Hereford e Shorthorn. Taxa de endogamia (ΔF), tamanho efetivo (Ne), grau de parentesco, entre outros parâmetros, foram estimados para fornecer subsídios aos programas de melhoramento. Os parâmetros indicaram adequada diversidade genética, com Ne variando entre 128 no Devon e 303 no Shorthorn e ΔF variando entre 1,50 no Angus e 3,92 no Devon; (2) avaliar estratégias de imputação de genótipos utilizando dados de Braford e Hereford através de painéis de baixa densidade (3K, 6K, 8K, 15K e 20K) para os painéis de 50K e 777K. Para o painel de 777K, também foram utilizados na imputação os painéis de 50K, 90iK e 90tK. Os resultados indicaram que, com exceção do painel de 3K, todos os demais painéis de baixa densidade poderiam ser utilizados como base visando à imputação para o painel de 50K e também que os painéis de média densidade (50K, 90iK e 90tK), poderiam ser utilizados como base na imputação para o painel de 777K. Esses painéis mostraram-se eficientes e possuem, em geral, custos compatíveis com a atividade pecuária; (3) avaliar a acurácia de predição dos valores genômicos utilizando alguns painéis de baixa densidade (8K e 15K) imputados para o painel de 50K, relacionando os resultados com o uso do painel original de 50K. A acurácia do valor genômico direto (DGV) e do valor genético genômico (GEBV) com o valor genético (EBV) utilizando painéis imputados ou não, indicaram que não houveram diferenças em acurácia e as perdas em acurácia por utilizar os painéis imputados ficaram entre -0,0002 e -0,0021 dependendo do painel, do cenário e da característica analisada; (4) usar marcadores moleculares na seleção genômica testando dois métodos BLUP (procedimento de passo único ou de multi passo) com dados simulados de bovinos de corte. Os resultados demonstraram, com base nos parâmetros estudados, igualdade de resultados entre os dois procedimentos; (5) avaliar a viabilidade do uso da seleção genômica usando dados de campo de animais Braford e Hereford, testando os dois métodos BLUP (passo único e multi passo). Os resultados com base nos parâmetros estudados, mostraram que as acurácias de predição do DGV e do GEBV foram iguais nos dois procedimentos, porém no método multi passo as predições genômicas foram menos viesadas. / The aim of this work were: (1) to evaluate parameters of genetic diversity and population structure based on Angus, Devon, Hereford and Shorthorn breeds. Inbreeding rate (ΔF), effective size (Ne), relatedness, among other parameters, were estimated to provide subsidies for breeding programs. The parameters indicated a good genetic diversity, with Ne ranging from 128 in Devon to 303 in Shorthorn and ΔF ranging from 1.50 in Angus to 3.92 in Devon; (2) to evaluate strategies of genotype imputation with Braford and Hereford beef data using low density panels (3K, 6K, 8K, 15K and 20K) for 50K and 777K panels. For imputation to the 777K panel were also used the 50K, 90iK and 90tK panels. The results indicated that, except for the 3K panel, all other low density panels could be used of imputation the 50K panel and also the medium density panels (50K, 90iK and 90tK), could be used of imputation the 777K panel. These panels have been efficient and have, in general, compatible costs of the beef cattle operation; (3) to evaluate the accuracy of genomic prediction using some low density panels (8K and 15K) imputed to the 50K panel, relating the results with 50K original panel. The accuracy of direct genomic value (DGV) and genomic estimated breeding value (GEBV) with estimated breeding value (EBV) using imputed or not panels, indicated that there were no differences in accuracy and the losses in accuracy by using the imputed panels ranged from -0.0002 to -0.0021 depending on the panel, the scenario and the trait; (4) to use molecular markers in genomic selection testing two BLUP methods (single and two steps) with simulated beef cattle data. The results showed, based on the parameters studied, equality of results between the two methods; (5) to evaluate the viability of using the genomic selection using Braford and Hereford beef cattle and testing the two BLUP methods (single and two steps). The results, based on the parameters studied, showed that DGV and GEBV accuracies were similar in both methods, but the genomic predictions were less biased with then two-step method.

A crescente preocupação com o meio ambiente pressiona a sociedade como um todo para a uma mudança rumo a hábitos mais sustentáveis. No setor produtivo, o impulso se dá pelo desenvolvimento de técnicas mais eficientes de produção, embasados em pesquisas e experimentos de campo. No setor florestal, além da preocupação com a técnicas de manejo e com o solo, o principal recurso a ser preservado é a água. Por meio do monitoramento de rios em bacias hidrográficas, séries históricas são coletadas, possibilitando o uso da teoria de séries temporais para ajuste de modelos pela metodologia Box e Jenkins. Em casos de monitoramentos de microbacias pareadas, existe a possibilidade de se comparar séries temporais, como descrito no presente trabalho. Em duas microbacias pareadas localizadas na região centro-leste do estado do Paraná, em uma fazenda no município de Telêmaco Borba, dados correspondendo a duas séries temporais distintas de vazão específica foram coletados. Devido a presença de falhas nos conjuntos de dados, uma metodologia para imputação foi utilizada de duas maneiras diferentes, possibilitando a posterior comparação das duas séries temporais pela metodologia de séries temporais. De acordo com os resultados, verifica-se que ambas as séries são diferentes tanto para o teste de comparação das funções de autocorrelação, quanto para o teste de comparação de séries temporais proposto por Silva, Ferreira e Sáfadi (2000). Portanto, segundo a caracterização dos estudos em microbacias pareadas, pode-se constatar que o manejo florestal empregado nos dois locais influenciam de forma diferente no comportamento da variável avaliada. / The growing concern for the enviroment presses society as a whole for a change towards sustainable habits. Regarding the production systems, more efficient production techniques based on research and field experiments are needed. As for forestry, besides the concern with management techniques and with soil preparation, the main resource to be preserved is water. Time series are collected by monitoring rivers in drainage basins, making possible the use of time series theory for fitting models based on Box and Jenkins methodology. When studying paired drainage basins, it is possible to compare time series, as described in this work. Two time series consisting of specific flow data were collected in a farm situated in the municipality of Telêmaco Borba, Eastern Paraná state, in two paired drainage basins. Because there were missing data, imputation techniques were used, making it possible to compare the two time series. Results showed that the time series are different for the comparison of the autocorrelation test and the time series comparison test proposed by Silva, Ferreira e Sáfadi (2000). Therefore, according to studies involving paired drainage basins, different forest management techniques influence differently the behavior of the response variable in the different drainage basins.

Prediction of energy load profiles is an important topic in Smart Grid technologies. Accurate forecasts can lead to reduced costs and decreased dependency on commercial power suppliers by adapting to prices on energy market, efficient utilisation of solar and wind energy and sophisticated load scheduling. This thesis compares various statistical and machine learning models and their ability to forecast load profile for an entire day divided into 48 half-hour intervals. Additionally, we examine various preprocessing methods and their influence on the accuracy of the models. We also compare a variety of imputation methods that are designed to reconstruct missing observation commonly present in energy consumption data.

隨著網路普及，人們越來越常在網路上購物並在線上評價商品，產生了非常大的口碑效應。不論對廠商或對消費者來說，線上商品評論都已經變得非常重要；消費者能藉由他人購買經驗判斷產品優劣，廠商能藉由消費者評價來提升產品品質，目前已有許多電子商務網站都有蒐集消費者購買產品後的意見回饋。 這些網站中有些提供消費者能對產品打一個總分並寫一段文字評論，然而每個消費者所評論的產品特徵通常各有不同，尤其是較晚購買的消費者更可能因為自己的意見已經有人提過而省略。將每個人提到的文字敘述量化為數字分數時，沒有寫到的特徵將會使量化後的資料存在許多遺漏值。 同時消費者也有可能提到一些不重要的特徵，若能找到消費者評論中，各個特徵影響消費者的多寡，廠商就能針對產品較重要的缺點改進。本研究將會著重探討消費者所提到的特徵對產品總分的影響，以及這些遺漏值填補後是否能接近消費者真實意見。 過去許多填補遺漏值的方法都是一次填補全部資料，並沒有考慮消費者會受到時間較早的評論影響。本研究設計一套多重插補的方法並透過模擬驗證，以之填補亞馬遜網站的Canon 系列 SX210、SX230、SX260等三個世代數位相機之消費者評論資料。研究結果指出此方法能夠準確估計各項特徵對產品總分的影響。 / Online user-generated product reviews have become a rich source of product quality information for both producers and customers. As a result, many E-commerce websites allow customers to rate products using scores, and some together with text comments. However, people usually comment only on the features they care about and might omit those have been mentioned by previous customers. Consequently, missing data occur when analyzing comments. In addition, customers may comment the features which influence neither their satisfaction nor sales volume. Thus, it is important to find the significant features so that manufacturers can improve the main defects. Our research focuses on modeling customer reviews and their influence on predicting overall ratings. We aim to understand whether, by filling up missing values, the critical features can be identified and the features rating authentically reflect customer opinion. Many previous studies fill whole the dataset, but not consider that customer reviews might be influenced by the foregoing reviews. We propose a method based on multiple imputation and fill the costumer reviews of Canon digital camera (SX210, SX230, SX260 generations) on Amazon. We design a simulation to verify the method’s effectiveness and the method get a great result on identifying the critical features.