Submitted by Ronildo Prado (ronisp@ufscar.br) on 2017-08-17T14:39:42Z No. of bitstreams: 1 DissTRR.pdf: 2739559 bytes, checksum: 80c76b7b0d4fcf15e1c9962556cd8745 (MD5) / Approved for entry into archive by Ronildo Prado (ronisp@ufscar.br) on 2017-08-17T14:39:52Z (GMT) No. of bitstreams: 1 DissTRR.pdf: 2739559 bytes, checksum: 80c76b7b0d4fcf15e1c9962556cd8745 (MD5) / Approved for entry into archive by Ronildo Prado (ronisp@ufscar.br) on 2017-08-17T14:39:58Z (GMT) No. of bitstreams: 1 DissTRR.pdf: 2739559 bytes, checksum: 80c76b7b0d4fcf15e1c9962556cd8745 (MD5) / Made available in DSpace on 2017-08-17T14:40:04Z (GMT). No. of bitstreams: 1 DissTRR.pdf: 2739559 bytes, checksum: 80c76b7b0d4fcf15e1c9962556cd8745 (MD5) Previous issue date: 2017-03-31 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / The frailty models are used to model the possible associations between survival times. Another alternative developed for modeling the dependence between multivariate data is the use of models based on copulas functions. In this paper we propose two derived survival models of copula of the Ali-Mikhail-Haq (AMH) and of the Frank to model the dependence of bivariate data in the presence of covariates and censored observations. For inferential purposes, we conducted a Bayesian approach using Monte Carlo methods in Markov Chain (MCMC). Some discussions on the model selection criteria were presented. In order to detect influential observations we use the Bayesian method of cases of deletion of influence analysis based on the difference ^. Finally, we show the applicability of the proposed models to sets of simulated and real data. We present, too, a new survival model with bivariate fraction of healing, which takes into account three settings for the latent activation mechanism: random activation, first activation and final activation. We apply this model to a set of Direct Credit loan data to the Consumer mode (DCC) and compare the settings, through Bayesian criteria for selection of models, which of the three models best fit. Finally, we show our future proposal for further research. / Os modelos de fragilidade são utilizados para modelar as possíveis associações entre os tempos de sobrevivência. Uma outra alternativa desenvolvida para modelar a dependência entre dados multivariados e o uso dos modelos baseados em funções cápulas. Neste trabalho propusemos dois modelos de sobrevivência derivados das copulas de Ali-Mikhail-Haq (AMH) e de Frank para modelar a dependência de dados bivariados na presença de covariáveis e observações censuradas. Para fins inferenciais, realizamos uma abordagem bayesiana usando métodos Monte Carlo em Cadeias de Markov (MCMC). Algumas discussões sobre os critérios de seleção de modelos são apresentadas. Com o objetivo de detectar observações influentes utilizamos o método bayesiano de analise de influencia de deleção de casos baseado na divergência. Por fim, mostramos a aplicabilidade dos modelos propostos a conjuntos de dados simulados e reais. Apresentamos, também, um novo modelo de sobrevivência bivariado com fração de cura, que leva em consideração três configurações para o mecanismo de ativação latente: ativação aleatória, primeira ativação e áltima ativação. Aplicamos este modelo a um conjunto de dados de empréstimo de Credito Direto ao modo do Consumidor (DCC) e comparamos os ajustes por meio dos critérios bayesianos de seleção de modelos para verificar qual dos três modelos melhor se ajustou. Por fim, mostramos nossa proposta futura para a continuaçaão da pesquisa.

The low grades that students score in some statistical units in Kenyan universities is of great concern and has evoked research interest in the teaching of some of the units and the students’ learning of the statistical content. The aim of the study was to investigate the difficulties undergraduate students experience in the learning of bivariate normal distribution in a Kenyan university. The research also aimed to answer the following research questions on the difficulties undergraduate students encounter in the learning of bivariate normal distribution. The first research question was based on the reasons why students find learning of bivariate normal distribution difficult and the second research question was to find the reasons why students experience such difficulties in learning bivariate normal distribution. The target population for this study included lecturers teaching statistics in the university, and second- and third- year students enrolled or who have previously completed the probability and statistics III unit, where the bivariate normal distribution content is covered. In selecting students for the study, the simple random sampling technique was employed while convenient sampling was used to select lecturers who participated in the study. A mixed methods design was adopted for this study where both quantitative and qualitative data was collected. A total of 175 students and six lecturers participated in this research study. All students who participated in the study did a bivariate normal distribution test (Appendix 1) designed by the researcher and then filled in a questionnaire (Appendix 2). The lecturers who participated in the study filled in an open-ended questionnaire (Appendix 3). The results showed that undergraduate students have difficulties in learning bivariate normal distribution. This is because most of them could neither state the bivariate normal distribution nor solve any of the application questions on the content. The students find it difficult to learn and comprehend the bivariate normal distribution equation with its many parameters and constants of the two random independent variables. The results also showed that students could not state the normal distribution equation nor could they solve questions on the normal distribution, which forms the foundational knowledge required for effective learning of the bivariate normal distribution content. ii Based on the results, the study recommended that emphasis should be placed on the basic and foundational knowledge of the normal distribution content and its applications before teaching bivariate normal distribution in probability and statistics III. In addition, it is recommended that all students should be involved in the learning of basic content to enable them to understand all parameters and constants in the equations and their applications. The study also recommends that lecturers revise the foundational knowledge and content related to the bivariate normal distribution before introducing and teaching the bivariate normal distribution content. This study also recommends that the university should consider a change of curriculum by teaching the bivariate normal distribution, as an introductory course to the unit under the multivariate distributions in statistics, in third year of the students’ studies. ; ; / Mathematics Education / M. Sc. (Mathematics Education)

Made available in DSpace on 2016-06-02T20:05:59Z (GMT). No. of bitstreams: 1 DissAKS.pdf: 566811 bytes, checksum: b01be331b665ab0824c5ab32218e4354 (MD5) Previous issue date: 2007-06-27 / Financiadora de Estudos e Projetos / The basic result of a soccer game is the final scoreboard, which can be seen as a bivariate random vector. Theoretically and based on existent literature we can argue that the number of marked gols by a team in a game obeys a (univariate) Poisson distribution. Thus, the Bivariate Poisson distributions are studied, in special for the class "of Holgate" (1964). Using as information the recent results of the teams, whose confrontation we want to model, several methods were used for parameters estimation of the Bivariate Poisson class "of Holgate". The idea is to use procedures that supply the probabilities of occurrence of placares, so that thus the probability of the occurrence of a certain result (team home´s victory, draw or defeat) can be calculated properly. The parameters of Bivariate Poisson distribution "of Holgate" are assumed to have a dependence factors, such as attack, defense and field, that possibly explain the numbers of goals. / O resultado básico de uma partida de futebol é o seu placar …nal, que pode ser visto como um vetor aleatório bivariado. Teoricamente e baseando-se na literatura existente podemos argumentar que o número de gols marcados por um time em uma dada partida obedeça a uma distribuição (univariada) de Poisson. Assim, são estudadas as distribuições de Poisson Bivariadas, com destaque para a classe "de Holgate" (1964). Utilizando como informações os resultados recentes dos times, cujo confronto se queira modelar, foram utilizados vários métodos para a estimação de parâmetros da densidade da classe Poisson Bivariada "de Holgate". A idéia é considerar procedimentos que forneçam as probabilidades de ocorrência de placares, para que assim a probabilidade da ocorrência de um determinado resultado (vitória do time mandante, empate ou derrota) possa ser obtido. Os parâmetros da distribuição de Poisson Bivariada "de Holgate" são assumidos terem dependência de fatores, tais como ataque, defesa e campo, que possivelmente explicam os números de gols feitos.

Made available in DSpace on 2016-06-02T20:06:04Z (GMT). No. of bitstreams: 1 3023.pdf: 2559668 bytes, checksum: 9cf8ca3c2627a6f2d69856b231e8a0a4 (MD5) Previous issue date: 2010-04-23 / Financiadora de Estudos e Projetos / In this dissertation we describe and implement procedures for nonparametric estimation of copulas and Sibuya function, and also procedures for bivariate analysis of dependence based on the behavior of their contours plot. Besisdes, we describe and implement the chiplot procedure and as well as a procedure for analising bivariate dependence in presence of censoring in the sample. Particularly, we propose a way to use it in a local correlation analysis. The performance of the proposed procedures are illustrated and evaluated in cases of very simple correlation, but also in a more complex correlation schemes. / Nesta dissertação descrevemos e implementamos procedimentos para estimação paramétrica da cópula e da função de Sibuya, e também procedimentos para análise de dependência bivariada, baseados no comportamento das suas curvas de nível. Também, descrevemos e implementamos o procedimento chi-plot e um procedimento para a análise de dependência bivariada com presença de censura na amostra. Particularmente, propomos formas de usá-los em análise de correlação local. O desempenho dos procedimentos propostos são ilustrados e avaliados em casos de estruturas de correlação simples, mas também em esquemas de correlação mais complexa.

Esta tese trata de dois assuntos importantes na teoria de risco: o fenômeno da dependência local e a identificação e mensuração de assimetrias apresentadas pelos dados. A primeira parte trata de dependência local, sendo abordadas algumas medidas já analisadas na literatura. Versões locais dos coeficientes de Kendall e Spearman , baseadas na distribuição condicional dos dados, são propostas. São apresentadas algumas propriedades dessas medidas e a aplicação das mesmas a algumas cópulas. Na segunda parte são apresentados resultados sobre cópulas bivariadas que são as menos associativas e menos bi-simétricas segundo o critério de máxima distância modular. A última parte trata da não-permutabilidade e assimetria radial dos dados. Uma medida de não-permutabilidade baseada nos coeficientes de correlação condicional é proposta e aplicada a algumas distribuições. No final, o conceito de quantil bivariado é aplicado nas definições de medidas para avaliar o grau de permutabilidade e de simetria radial presentes na estrutura de dependência dos dados e de testes de hipóteses para verificar se a cópula subjacente aos dados é permutável ou radialmente simétrica. / In this thesis two important fields in risk theory are studied: the local dependence phenomenon and the identification and measuring of asymmetries contained in data. The first part deals with local dependence: some measures already studied in the literature are presented and discussed, and local versions of the coefficients Kendall and Spearman , based on the conditional distribution of data, are proposed. Properties of these measures and some examples concerning its application are treated. In the second part are presented some results about bivariate copulas which are the least associative and the least bi-symmetric according to the maximum modular distance. The last part analyses the nonexchangeability and the radial asymmetry of data. A measure of nonexchangeability based on the conditional correlation coefficient is proposed and applied to some distribution functions. At the end, the concept of bivariate quantile is applied in the definitions of measures for evaluating the degree of exchangeability and radial symmetry present in data and of hypothesis tests proposed for verifying whether the underlying copula is exchangeable or radially symmetric.

Submitted by Luciana Sebin (lusebin@ufscar.br) on 2016-09-20T17:56:33Z No. of bitstreams: 1 TeseWLO.pdf: 4950921 bytes, checksum: adf4e51b260eb25ef34f14a351022af5 (MD5) / Approved for entry into archive by Marina Freitas (marinapf@ufscar.br) on 2016-09-21T12:48:10Z (GMT) No. of bitstreams: 1 TeseWLO.pdf: 4950921 bytes, checksum: adf4e51b260eb25ef34f14a351022af5 (MD5) / Approved for entry into archive by Marina Freitas (marinapf@ufscar.br) on 2016-09-21T12:48:16Z (GMT) No. of bitstreams: 1 TeseWLO.pdf: 4950921 bytes, checksum: adf4e51b260eb25ef34f14a351022af5 (MD5) / Made available in DSpace on 2016-09-21T12:48:30Z (GMT). No. of bitstreams: 1 TeseWLO.pdf: 4950921 bytes, checksum: adf4e51b260eb25ef34f14a351022af5 (MD5) Previous issue date: 2016-01-28 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / In this thesis, a wide general class of models for mixed responses is proposed in which joint distributions are constructed by the conditional approach (probability density functions, (pdf), as the product of a marginal pdf and a conditional pdf). It is assumed that the distribution of the discrete response and the conditional distribution of the continuous response given the discrete variable belong to one- or two-parameter exponential family of distributions. Furthermore, the marginal means are related to the covariates by link functions using linear and/or nonlinear and/or non-parametric predictors and a dependency structure between the responses is introduced into the model via the conditional mean. Estimation methods, diagnostic analysis and in uence techniques are presented as well as a simulation study considering the Bernoulli-exponential and Poisson-normal semiparametric models, two particular cases of the proposed class. Finally, one of the proposed models is used in a real data set involving the total cost of care for each patient during hospitalization, the use or not of the intensive treatment units and the age of the patient. / Nesta tese é proposta uma classe ampla e geral de modelos bivariados para respostas mistas em que as distribuições conjuntas são construídas pelo método da fatoração (função densidade de probabilidade, (fdp), como o produto de uma fdp marginal e uma fdp condicional). É assumido que a distribuição da variável resposta discreta e a distribuição condicional da variável resposta contínua dada a variável discreta pertencem à família exponencial de distribuições uniparamétrica ou biparamétrica. Além disso, as médias marginais são relacionadas a covariáveis através de funções de ligação usando preditores linear e/ou não linear e/ou não paramétrico e uma estrutura de dependência entre as respostas é inserida no modelo via a média condicional. Métodos de estimação, análises de diagnóstico e técnicas de in uência são apresentadas assim como um estudo de simulação considerando os modelos Bernoulli-exponencial e Poisson-normal semiparamétrico, dois casos particulares da classe proposta. Finalmente, um dos modelos propostos é usado em um conjunto de dados reais envolvendo gastos totais com cuidados para cada paciente durante a hospitalização, o uso ou não da unidade de tratamento intensivo e a idade do paciente.

Neste trabalho são reunidos diferentes modelos e técnicas para representar situações experimentais ou observacionais de análise de sobrevivência. Para modelar respostas bivariadas e covariáveis foi proposto o modelo de regressão Kumaraswamy-Weibull bivariado. A presen»ca de indivíduos curados foi considerada sob duas diferentes abordagens, originando o modelo de regressão com fração de cura para dados bivariados por meio de cópulas e o modelo de regressão log-linear bivariado com fração de cura. Os parâmetros dos modelos foram esti- mados pelo método de máxima verossimilhança sujeito a restriçãoo nos parâmetros por meio da função barreira adaptada. Adaptou-se uma análise de sensibilidade de forma a considerar as metodologias de Influência Global, Influência Local e Influência Local Total para verificar vários aspectos que envolvem a formulação e ajuste dos modelos propostos. Utilizou-se um conjunto de dados de insuficiência renal e retinopatia diabética são utilizados para exemplificar a aplicação dos modelos propostos. / This work brought together di®erent models and techniques to represent expe- rimental or observational situations in survival analysis. To model bivariate responses and covariates was proposed Kumaraswamy Weibull bivariate regression model. The presence of cured individuals was considered under two di®erent approaches originating the regression model with a cured fraction for bivariate data through copulas and the log-linear bivariate regression model with cured fraction. The parameters of the models were estimated by ma- ximum likelihood method subject to the restriction on the parameters through the adapted barrier function. A sensitivity analysis was adapted considering the methodologies of Global In°uence, Local In°uence and Total Local In°uence to check various aspects of the formulation and adjustment of the models proposed. Data set of renal failure and diabetic retinopathy are used to exemplify the application of the proposed models.

Made available in DSpace on 2017-07-10T19:23:53Z (GMT). No. of bitstreams: 1 protegidoJacqueline_dissertacao.pdf: 3012964 bytes, checksum: 337f371fa8c665bb4fdfdc22709938da (MD5) Previous issue date: 2015-02-02 / The soil spatial variability s studies are based in geostatistics which appears as a method whose data comes from natural phenomena and consider the geographical location of the phenomenon. If in an area under study the researcher has interest in searching the variability of variables and has evidence that the steps which describe the spatial structure of this variables aren t independent a geostatistical bivariate model study can be proposed. This work concentrates on evaluating the variation s influence of the bivariate Gaussian common component model (BGCCM) parameters in calculating the Pearson correlation coefficient and analyzing the influence that sizes and sampling settings may present at the BGCCM s estimation and at spatial prediction variables in non-sampled locations. Moreover, for co-placed samples, crossed semivariograms were built and compared with univariate model and BGCCM, in relation to estimates of the model and the sizes associated with the spatial prediction. In order to do it, these methodologies were applied in simulated data sets and experimental data, from an agricultural property. The simulation study of the parameters variation influence s analysis of the bivariate model BGCCM in calculating the Pearson correlation coefficient between described variables of bivariate model BGCCM revealed that the Pearson s linear correlation coefficient can t be considered in decision-making about the presence of joint spatial dependence between pairs of variables. In the study with simulations, it was observed that the biggest differences of accuracy measures and the square sum of the spatial prediction s difference occurred when the univariate models and crossed semivariogram were compared to the BGCCM. Moreover, the simulation s study observed that for balanced data the regular and irregular meshes showed better efficiency as the spatial prediction. The study with real data showed that under the BGCCM approach, spatial dependence was observed, mainly between pH and Mn for co-placed and balanced data of the agriculture year 2010/2011; and between the variable inside of the next pairs: (Prod, RSP0-10), (Prod, RSP11-20), (Prod, RSP21-30) and (Prod, Mn) for co-placed and balanced data of the agriculture year 2013/2014; and (Prod, RSP11-20) and (Prod, RSP21-30) for co-placed and unbalanced data of the agriculture year 2013/2014. Still considering the real data study comparing the univariated models, crossed semivariogram and BGCCM, differences could be observed in the square sum of prediction s difference and in the accuracy measures, both for balanced and unbalanced data. However, considering the real data and the control sample, the spatial prediction s quality using the BGCCM model was inefficient when compared to the quality resulted from the spatial prediction using the univariate model. Nevertheless, this result may have been influenced by the choice of sample configuration. / Estudos da variabilidade espacial do solo estão baseados na geoestatística, que se apresenta como um método cujos dados provêm de fenômenos naturais e que consideram a localização geográfica do fenômeno. Se numa área em estudo o pesquisador tem interesse em pesquisar a variabilidade das variáveis e se há evidências que os passos que descrevem a estrutura espacial dessas variáveis não são independentes pode-se propor o estudo de um modelo geoestatístico bivariado. Este trabalho concentrou-se em avaliar a influência da variação dos parâmetros do modelo bivariado com componente de correlação parcialmente comum (bivariate Gaussian common component model BGCCM) no cálculo do coeficiente de correlação linear de Pearson e analisar a influência que tamanhos e configurações amostrais podem apresentar na estimação do modelo BGCCM e na predição espacial de variáveis em localizações não amostradas. Além disso, para amostras co-locadas, construíram-se os semivariogramas cruzados e comparou-se com o modelo univariado e BGCCM, em relação às estimativas do modelo e as medidas associadas à predição espacial. Para isso, essas metodologias foram aplicadas em conjuntos de dados simulados e dados experimentais, provenientes de uma propriedade agrícola. O estudo de simulação da análise da influência da variação dos parâmetros do modelo bivariado BGCCM no cálculo do coeficiente de correlação linear de Pearson entre as variáveis descritas do modelo bivariado BGCCM revelou que o coeficiente de correlação linear de Pearson não pode ser considerado na tomada de decisão quanto à presença de dependência espacial conjunta entre pares de variáveis. No estudo com simulações observou-se que as maiores diferenças das medidas de acurácia e da soma quadrada da diferença entre as predições espaciais ocorreram quando se comparou os modelos univariado e semivariograma cruzado com o BGCCM. Ainda no estudo de simulação observou-se que para os dados balanceados as malhas regular e irregular apresentaram melhor eficiência quanto à predição espacial. O estudo com dados reais mostrou que, sob a abordagem do modelo BGCCM, observou-se a presença de dependência espacial principalmente entre pH e Mn para dados co-locados e balanceados do ano agrícola 2010/2011; e entre as variáveis dentro dos seguintes pares: (Prod, RSP0-10), (Prod, RSP11-20), (Prod, RSP21-30) e (Prod, Mn) para dados co-locados e balanceados do ano agrícola 2013/2014; e (Prod, RSP11-20) e (Prod, RSP21-30) para dados co-locados e desbalanceados do ano agrícola 2013/2014. Ainda considerando o estudo com dados reais comparando os modelos univariado, semivariograma cruzado e BGCCM, mostraram diferenças na soma quadrada da diferença da predição e nas medidas acurácia, tanto para dados balanceados como para os desbalanceados. No entanto, considerando os dados reais e a amostra controle, a qualidade da predição espacial usando o modelo BGCCM se mostrou ineficiente quando comparada com a qualidade obtida na predição espacial usando o modelo univariado. Porém, esse resultado pode ter sido influenciado pela escolha da configuração amostral utilizada.

Most albatrosses are critically endangered, endangered or vulnerable due to the deleterious impact of fisheries, pollution, introduced species, habitat alteration, and climate change. Foraging behaviour influences many aspects of seabird biology, and a detailed understanding of foraging ecology is required to better predict the impacts of significant changes to the marine environment. Campbell Albatross (Thalassarche impavida) is a threatened endemic, confined to a small number of locations on Campbell Island, New Zealand and was recently split from the closely related Black-browed Albatross (T. melanophrys). We currently lack much basic information on the foraging behaviour of this species, hindering our ability to understand how change may have occurred in the past and make predictions about it’s long-term future. First, I used GPS loggers and stable isotope analysis of blood to investigate how distribution and foraging effort (distance travelled and duration) varied with sex and breeding stage. I found that Campbell Albatrosses are sexually dimorphic and showed sex-specific foraging behaviour and habitat use – although this varied by stage of reproduction. Because males and females may be vulnerable to different threats, such as interactions with fisheries, I compared the spatial overlap and high resolution spatio-temporal overlaps between fisheries vessels and albatrosses within New Zealand’s Exclusive Economic Zone (EEZ). Albatrosses utilised 32% of the EEZ, however they overlapped with fisheries vessels in only 0.20% of the area. Previous research has demonstrated that the influence of fisheries vessels goes beyond the immediate location of the boat itself. Campbell Albatross have low levels of spatio-temporal overlap with fisheries – with males overlapping more than females. More generally, my results indicate that adding data on fine scale interactions will improve fisheries risk assessments, and provide information needed for the conservation and management of the Campbell Albatross. A key development in recent ecological research has been a greater appreciation that inter-individual variation in foraging behaviour can have profound population-level consequences. Accordingly I tested for individual differences in foraging behaviour in Campbell Albatrosses. The majority of individuals demonstrated both annual and inter-annual individual consistency in foraging locations, and the degree of specialisation was influenced by both sex and year. Consistent terminal latitude and longitude of foraging trips indicated high foraging area fidelity with a degree of flexibility in the fine-scale location. During brooding, females used the Campbell Plateau and showed more consistent behaviours than males, which tended to forage in the Southern Ocean. This adds to a growing body of evidence of individual foraging specialisation among seabirds in general and albatrosses in particular and reveals marked inter-individual differences in vulnerability to threats. In light of the evidence of individual foraging specialisations in the Campbell albatross, I also preformed a literature review of individual foraging specialisations across all seabirds. I found studies examining foraging specialisation for 35 species, with 28 (80%) providing evidence of consistent inter-individual differences (i.e. specialisation). Current studies suggest that specialisation is influenced by environmental variability and resource predictability, however, with limited data in tropical regions, more studies are needed to test these links. In summary, my thesis has provided new information on Campbell Albatross foraging ecology. Sex specific variations in behaviour and habitat use may influence conservation and management strategies. I have been able to contextualise the consistent individual differences in foraging distribution described for this species in light of global patterns of individual foraging specialisation in seabirds and highlight future areas of research.

Orientador: Filidor Edilfonso Vilca Labra / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Matemática Estatística e Computação Científica / Made available in DSpace on 2018-08-27T16:25:27Z (GMT). No. of bitstreams: 1 RamosQuispe_LuzMarina_M.pdf: 6411257 bytes, checksum: 6e1e798cf8f6d7586fe5d9a057492a77 (MD5) Previous issue date: 2015 / Resumo: Vários trabalhos têm sido feitos sobre a distribuição Birnbaum-Saunders (BS) univariada e suas extensões. A distribuição bivariada Birnbaum-Saunders (BS) foi apresentada apenas recentemente por Kundu et al. (2010) e algumas extensões já foram discutidas por Vilca et al. (2014) e Kundu et al. (2013). Eles propuseram uma distribuição BS bivariada com estrutura de dependência e estabeleceram várias propriedades atraentes. Este trabalho fornece extensões, univariada e bivariada, da distribuição BS. Estas extensões são baseadas na distribuição Gaussiana Inversa (IG) que é usada como uma distribuição de mistura no contexto de misturas de escala normal. As distribuições resultantes são distribuições absolutamente contínuas e muitas propriedades da distribuição BS são preservadas. Sob caso bivariado, as marginais e condicionais são do tipo Birnbaum-Saunders univariada. Para a obtenção da estimativa de máxima verossimilhança (EMV) é desenvolvido um algoritmo EM. Ilustramos os resultados obtidos com dados reais e simulados / Abstract: Several works have been done on the univariate Birnbaum-Saunders (BS) distribution and its extensions. The bivariate Birnbaum-Saunders (BS) distribution was presented only recently by Kundu et al. (2010) and some extensions have already been discussed by Vilca et al. (2014) and Kundu et al. (2013). They proposed a bivariate BS distribution with dependence structure and established several attractive properties. This work provides extensions, univariate and bivariate, of the BS distribution. These extensions are based on the Inverse Gaussian (IG) distribution that is used as a mixing distribution in the context of scale mixtures of normal. The resulting distributions are absolutely continuous distributions and many properties of the BS distribution are preserved. Under bivariate case, the marginals and conditionals are of type univariate Birnbaum-Saunders. For obtaining the maximum likelihood estimates (MLE) of the model parameters is developed an algorithm EM. We illustrate the obtained results with real and simulated dataset / Mestrado / Estatistica / Mestra em Estatística