Muitos equipamentos utilizados para quantificar substâncias, como toxinas em alimentos, freqüentemente apresentam deficiências para quantificar quantidades baixas. Em tais casos, geralmente indicam a ausência da substância quando esta existe, mas está abaixo de um valor pequeno \'ksi\' predeterminado, produzindo valores iguais a zero não necessariamente verdadeiros. Em outros casos, detectam a presença da substância, mas são incapazes de quantificá-la quando a quantidade da substância está entre \'ksai\' e um valor limiar \'tau\', conhecidos. Por outro lado, quantidades acima desse valor limiar são quantificadas de forma contínua, dando origem a uma variável aleatória contínua X cujo domínio pode ser escrito como a união dos intervalos, [ómicron, \"ksai\'), [\"ksai\', \'tau\' ] e (\'tau\', ?), sendo comum o excesso de valores iguais a zero. Neste trabalho, são propostos modelos que possibilitam discriminar a probabilidade de zeros verdadeiros, como o modelo de mistura com dois componentes, sendo um degenerado em zero e outro com distribuição contínua, sendo aqui consideradas as distribuições: exponencial, de Weibull e gama. Em seguida, para cada modelo, foram observadas suas características, propostos procedimentos para estimação de seus parâmetros e avaliados seus potenciais de ajuste por meio de métodos de simulação. Finalmente, a metodologia desenvolvida foi ilustrada por meio da modelagem de medidas de contaminação com aflatoxina B1, observadas em grãos de milho, de três subamostras de um lote de milho, analisados no Laboratório de Micotoxinas do Departamento de Agroindústria, Alimentos e Nutrição da ESALQ/USP. Como conclusões, na maioria dos casos, as simulações indicaram eficiência dos métodos propostos para as estimações dos parâmetros dos modelos, principalmente para a estimativa do parâmetro \'delta\' e do valor esperado, \'Epsilon\' (Y). A modelagem das medidas de aflatoxina, por sua vez, mostrou que os modelos propostos são adequados aos dados reais, sendo que o modelo de mistura com distribuição de Weibull, entretanto, ajustou-se melhor aos dados. / Much equipment used to quantify substances, such as toxins in foods, is unable to measure low amounts. In cases where the substance exists, but in an amount below a small fixed value \'ksi\' , the equipment usually indicates that the substance is not present, producing values equal to zero. In cases where the quantity is between \'\'ksi\' and a known threshold value \'tau\', it detects the presence of the substance but is unable to measure the amount. When the substance exists in amounts above the threshold value ?, it is measure continuously, giving rise to a continuous random variable X whose domain can be written as the union of intervals, [ómicron, \"ksai\'), [\"ksai\', \'tau\' ] and (\'tau\', ?), This random variable commonly has an excess of zero values. In this work we propose models that can detect the probability of true zero, such as the mixture model with two components, one being degenerate at zero and the other with continuous distribution, where we considered the distributions: exponential, Weibull and gamma. Then, for each model, its characteristics were observed, procedures for estimating its parameters were proposed and its potential for adjustment by simulation methods was evaluated. Finally, the methodology was illustrated by modeling measures of contamination with aflatoxin B1, detected in grains of corn from three sub-samples of a batch of corn analyzed at the laboratory of of Mycotoxins, Department of Agribusiness, Food and Nutrition ESALQ/USP. In conclusion, in the majority of cases the simulations indicated that the proposed methods are efficient in estimating the parameters of the models, in particular for estimating the parameter ? and the expected value, E(Y). The modeling of measures of aflatoxin, in turn, showed that the proposed models are appropriate for the actual data, however the mixture model with a Weibull distribution fits the data best.

<p>Recently, the Conway-Maxwell Poisson (COM-Poisson) cure rate model has been proposed which includes as special cases some of the well-known cure rate models discussed in the literature. Data obtained from cancer clinical trials are often right censored and the expectation maximization (EM) algorithm can be efficiently used for the determination of the maximum likelihood estimates (MLEs) of the model parameters based on right censored data.</p> <p>By assuming the lifetime distribution to be exponential, lognormal, Weibull, and gamma, the necessary steps of the EM algorithm are developed for the COM-Poisson cure rate model and some of its special cases. The inferential method is examined by means of an extensive simulation study. Model discrimination within the COM-Poisson family is carried out by likelihood ratio test as well as by information-based criteria. Finally, the proposed method is illustrated with a cutaneous melanoma data on cancer recurrence. As the lifetime distributions considered are not nested, it is not possible to carry out a formal statistical test to determine which among these provides an adequate fit to the data. For this reason, the wider class of generalized gamma distributions is considered which contains all of the above mentioned lifetime distributions as special cases. The steps of the EM algorithm are then developed for this general class of distributions and a simulation study is carried out to evaluate the performance of the proposed estimation method. Model discrimination within the generalized gamma family is carried out by likelihood ratio test and information-based criteria. Finally, for the considered cutaneous melanoma data, the two-way flexibility of the COM-Poisson family and the generalized gamma family is utilized to carry out a two-way model discrimination to select a parsimonious competing cause distribution along with a suitable choice of a lifetime distribution that provides the best fit to the data.</p> / Doctor of Philosophy (PhD)

Propor uma família de distribuição de probabilidade mais ampla e flexível é de grande importância em estudos estatísticos. Neste trabalho é utilizado um novo método de adicionar um parâmetro para uma distribuição contínua. A distribuição gama generalizada, que tem como casos especiais a distribuição Weibull, exponencial, gama, qui-quadrado, é usada como distribuição base. O novo modelo obtido tem quatro parâmetros e é chamado odd log-logística gama generalizada (OLLGG). Uma das características interessante do modelo OLLGG é o fato de apresentar bimodalidade. Outra proposta deste trabalho é introduzir um modelo de regressão chamado log-odd log-logística gama generalizada (LOLLGG) com base na GG (Stacy e Mihram, 1965). Este modelo pode ser muito útil, quando por exemplo, os dados amostrados possuem uma mistura de duas populações estatísticas. Outra vantagem da distribuição OLLGG consiste na capacidade de apresentar várias formas para a função de risco, crescente, decrescente, na forma de U e bimodal entre outras. Desta forma, são apresentadas em ambos os casos as expressões explícitas para os momentos, função geradora e desvios médios. Considerando dados nãocensurados e censurados de forma aleatória, as estimativas para os parâmetros de interesse, foram obtidas via método da máxima verossimilhança. Estudos de simulação, considerando diferentes valores para os parâmetros, porcentagens de censura e tamanhos amostrais foram conduzidos com o objetivo de verificar a flexibilidade da distribuição e a adequabilidade dos resíduos no modelo de regressão. Para ilustrar, são realizadas aplicações em conjuntos de dados reais. / Providing a wider and more flexible probability distribution family is of great importance in statistical studies. In this work a new method of adding a parameter to a continuous distribution is used. In this study the generalized gamma distribution (GG) is used as base distribution. The GG distribution has, as especial cases, Weibull distribution, exponential, gamma, chi-square, among others. For this motive, it is considered a flexible distribution in data modeling procedures. The new model obtained with four parameters is called log-odd log-logistic generalized gamma (OLLGG). One of the interesting characteristics of the OLLGG model is the fact that it presents bimodality. In addition, a regression model regression model called log-odd log-logistic generalized gamma (LOLLGG) based by GG (Stacy e Mihram, 1965) is introduced. This model can be very useful when, the sampled data has a mixture of two statistical populations. Another advantage of the OLLGG distribution is the ability to present various forms for the failing rate, as increasing, as decreasing, and the shapes of bathtub or U. Explicity expressions for the moments, generating functions, mean deviations are obtained. Considering non-censored and randomly censored data, the estimates for the parameters of interest were obtained using the maximum likelihood method. Simulation studies, considering different values for the parameters, percentages of censoring and sample sizes were done in order to verify the distribuition flexibility, and the residues distrbutuon in the regression model. To illustrate, some applications using real data sets are carried out.

Propor uma família de distribuição de probabilidade mais ampla e flexível é de grande importância em estudos estatísticos. Neste trabalho é utilizado um novo método de adicionar um parâmetro para uma distribuição contínua. A distribuição gama generalizada, que tem como casos especiais a distribuição Weibull, exponencial, gama, qui-quadrado, é usada como distribuição base. O novo modelo obtido tem quatro parâmetros e é chamado odd log-logística gama generalizada (OLLGG). Uma das características interessante do modelo OLLGG é o fato de apresentar bimodalidade. Outra proposta deste trabalho é introduzir um modelo de regressão chamado log-odd log-logística gama generalizada (LOLLGG) com base na GG (Stacy e Mihram, 1965). Este modelo pode ser muito útil, quando por exemplo, os dados amostrados possuem uma mistura de duas populações estatísticas. Outra vantagem da distribuição OLLGG consiste na capacidade de apresentar várias formas para a função de risco, crescente, decrescente, na forma de U e bimodal entre outras. Desta forma, são apresentadas em ambos os casos as expressões explícitas para os momentos, função geradora e desvios médios. Considerando dados nãocensurados e censurados de forma aleatória, as estimativas para os parâmetros de interesse, foram obtidas via método da máxima verossimilhança. Estudos de simulação, considerando diferentes valores para os parâmetros, porcentagens de censura e tamanhos amostrais foram conduzidos com o objetivo de verificar a flexibilidade da distribuição e a adequabilidade dos resíduos no modelo de regressão. Para ilustrar, são realizadas aplicações em conjuntos de dados reais. / Providing a wider and more flexible probability distribution family is of great importance in statistical studies. In this work a new method of adding a parameter to a continuous distribution is used. In this study the generalized gamma distribution (GG) is used as base distribution. The GG distribution has, as especial cases, Weibull distribution, exponential, gamma, chi-square, among others. For this motive, it is considered a flexible distribution in data modeling procedures. The new model obtained with four parameters is called log-odd log-logistic generalized gamma (OLLGG). One of the interesting characteristics of the OLLGG model is the fact that it presents bimodality. In addition, a regression model regression model called log-odd log-logistic generalized gamma (LOLLGG) based by GG (Stacy e Mihram, 1965) is introduced. This model can be very useful when, the sampled data has a mixture of two statistical populations. Another advantage of the OLLGG distribution is the ability to present various forms for the failing rate, as increasing, as decreasing, and the shapes of bathtub or U. Explicity expressions for the moments, generating functions, mean deviations are obtained. Considering non-censored and randomly censored data, the estimates for the parameters of interest were obtained using the maximum likelihood method. Simulation studies, considering different values for the parameters, percentages of censoring and sample sizes were done in order to verify the distribuition flexibility, and the residues distrbutuon in the regression model. To illustrate, some applications using real data sets are carried out.

In Paper [I] we use data on Swedish public procurement auctions for internal regularcleaning service contracts to provide novel empirical evidence regarding green publicprocurement (GPP) and its effect on the potential suppliers’ decision to submit a bid andtheir probability of being qualified for supplier selection. We find only a weak effect onsupplier behavior which suggests that GPP does not live up to its political expectations.However, several environmental criteria appear to be associated with increased complexity,as indicated by the reduced probability of a bid being qualified in the postqualificationprocess. As such, GPP appears to have limited or no potential to function as an environmentalpolicy instrument. In Paper [II] the observation is made that empirical evaluations of the effect of policiestransmitted through public procurements on bid sizes are made using linear regressionsor by more involved non-linear structural models. The aspiration is typically to determinea marginal effect. Here, I compare marginal effects generated under both types ofspecifications. I study how a political initiative to make firms less environmentally damagingimplemented through public procurement influences Swedish firms’ behavior. Thecollected evidence brings about a statistically as well as economically significant effect onfirms’ bids and costs. Paper [III] embarks by noting that auction theory suggests that as the number of bidders(competition) increases, the sizes of the participants’ bids decrease. An issue in theempirical literature on auctions is which measurement(s) of competition to use. Utilizinga dataset on public procurements containing measurements on both the actual and potentialnumber of bidders I find that a workhorse model of public procurements is bestfitted to data using only actual bidders as measurement for competition. Acknowledgingthat all measurements of competition may be erroneous, I propose an instrumental variableestimator that (given my data) brings about a competition effect bounded by thosegenerated by specifications using the actual and potential number of bidders, respectively.Also, some asymptotic results are provided for non-linear least squares estimatorsobtained from a dependent variable transformation model. Paper [VI] introduces a novel method to measure bidders’ costs (valuations) in descending(ascending) auctions. Based on two bounded rationality constraints bidders’costs (valuations) are given an imperfect measurements interpretation robust to behavioraldeviations from traditional rationality assumptions. Theory provides no guidanceas to the shape of the cost (valuation) distributions while empirical evidence suggeststhem to be positively skew. Consequently, a flexible distribution is employed in an imperfectmeasurements framework. An illustration of the proposed method on Swedishpublic procurement data is provided along with a comparison to a traditional BayesianNash Equilibrium approach.

Stacy (1962)首先提出廣義伽瑪分配 (generalized gamma distribution)，此分布被廣泛應用於存活分析 (survival analysis) 以及可靠度 (reliability) 中壽命時間的資料描述。事實上，像是指數分配 (exponential distribution)、韋伯分配 (Weibull distribution) 以及伽瑪分配 (gamma distribution) 都是廣義伽瑪分配的一個特例。 Bologna (1987)提出一個特殊的雙變量廣義伽瑪分配 (bivariate generalized gamma distribution) 可以經由雙變量常態分配 (bivariate normal distribution) 所推得。我們根據他的想法，提出多變量廣義伽瑪分配可以經由多變量常態分配所推得。在過去的研究中，學者們做了許多有關雙變量伽瑪分配。當我們提到雙變量常態分配，由於其分配的型式為唯一的，所以沒人任何人對其分配的型式有疑問。然而，雙變量伽瑪分配卻有很多不同的型式。 在這篇論文中的架構如下。在第二章中，我們介紹並討論雙變量廣義伽瑪分配可以經由雙變量常態分配所推得，接著推導參數估計以及介紹模擬的程序。在第三章中，我們介紹一些對稱以及非對稱的雙變量伽瑪分配，接著拓展到雙變量廣義伽瑪分配，有關參數的估計以及模擬結果也將在此章中討論。在第三章最後，我們建構參數的敏感度分析 (sensitivity analysis)。最後，在第四章中，我們陳述結論以及未來研究方向。 / The generalized gamma distribution was introduced by Stacy (1962). This distribution is useful to describe lifetime data when conducting survival analysis and reliability. In fact, it includes the widely used exponential, Weibull, and gamma distributions as special cases. Bologna (1987) showed that a special bivariate genenralized gamma distribution can be derived from a bivariate normal distribution. Follow his idea, we show that a multivariate generalized gamma distribution can be derived from a multivariate normal distribution. In the past, researchers spend much time in working on a bivariate gamma distribution. When a bivariate normal distribution is mentioned, no one feels puzzled about its form, since it has only one form. However, there are various forms of bivariate gamma distributions. In this paper is as following. In Chapter 2, we introduce and discuss the bivariate generalized gamma distribution, then the multivariate generalized gamma distribution is derived. We also develop parameters estimation and simulation procedure. In Chapter 3, we introduce some symmetrical and asymmetrical bivariate gamma distributions, then they are extended to the bivariate generalized gamma distributions. Problems of parameters estimation and simulation results are also discussed in Chapter 3. Besides, sensitivity analyses of parameters estimation are conducted. Finally, we state conclusion and future work in Chapter 4.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal.

Les titres financiers sont souvent modélisés par des équations différentielles stochastiques (ÉDS). Ces équations peuvent décrire le comportement de l'actif, et aussi parfois certains paramètres du modèle. Par exemple, le modèle de Heston (1993), qui s'inscrit dans la catégorie des modèles à volatilité stochastique, décrit le comportement de l'actif et de la variance de ce dernier. Le modèle de Heston est très intéressant puisqu'il admet des formules semi-analytiques pour certains produits dérivés, ainsi qu'un certain réalisme. Cependant, la plupart des algorithmes de simulation pour ce modèle font face à quelques problèmes lorsque la condition de Feller (1951) n'est pas respectée. Dans ce mémoire, nous introduisons trois nouveaux algorithmes de simulation pour le modèle de Heston. Ces nouveaux algorithmes visent à accélérer le célèbre algorithme de Broadie et Kaya (2006); pour ce faire, nous utiliserons, entre autres, des méthodes de Monte Carlo par chaînes de Markov (MCMC) et des approximations. Dans le premier algorithme, nous modifions la seconde étape de la méthode de Broadie et Kaya afin de l'accélérer. Alors, au lieu d'utiliser la méthode de Newton du second ordre et l'approche d'inversion, nous utilisons l'algorithme de Metropolis-Hastings (voir Hastings (1970)). Le second algorithme est une amélioration du premier. Au lieu d'utiliser la vraie densité de la variance intégrée, nous utilisons l'approximation de Smith (2007). Cette amélioration diminue la dimension de l'équation caractéristique et accélère l'algorithme. Notre dernier algorithme n'est pas basé sur une méthode MCMC. Cependant, nous essayons toujours d'accélérer la seconde étape de la méthode de Broadie et Kaya (2006). Afin de réussir ceci, nous utilisons une variable aléatoire gamma dont les moments sont appariés à la vraie variable aléatoire de la variance intégrée par rapport au temps. Selon Stewart et al. (2007), il est possible d'approximer une convolution de variables aléatoires gamma (qui ressemble beaucoup à la représentation donnée par Glasserman et Kim (2008) si le pas de temps est petit) par une simple variable aléatoire gamma. / Financial stocks are often modeled by stochastic differential equations (SDEs). These equations could describe the behavior of the underlying asset as well as some of the model's parameters. For example, the Heston (1993) model, which is a stochastic volatility model, describes the behavior of the stock and the variance of the latter. The Heston model is very interesting since it has semi-closed formulas for some derivatives, and it is quite realistic. However, many simulation schemes for this model have problems when the Feller (1951) condition is violated. In this thesis, we introduce new simulation schemes to simulate price paths using the Heston model. These new algorithms are based on Broadie and Kaya's (2006) method. In order to increase the speed of the exact scheme of Broadie and Kaya, we use, among other things, Markov chains Monte Carlo (MCMC) algorithms and some well-chosen approximations. In our first algorithm, we modify the second step of the Broadie and Kaya's method in order to get faster schemes. Instead of using the second-order Newton method coupled with the inversion approach, we use a Metropolis-Hastings algorithm. The second algorithm is a small improvement of our latter scheme. Instead of using the real integrated variance over time p.d.f., we use Smith's (2007) approximation. This helps us decrease the dimension of our problem (from three to two). Our last algorithm is not based on MCMC methods. However, we still try to speed up the second step of Broadie and Kaya. In order to achieve this, we use a moment-matched gamma random variable. According to Stewart et al. (2007), it is possible to approximate a complex gamma convolution (somewhat near the representation given by Glasserman and Kim (2008) when T-t is close to zero) by a gamma distribution.

Les radars à synthèse d’ouverture (RSO) polarimétriques sont devenus incontournables dans le domaine de la télédétection, grâce à leur zone de couverture étendue, ainsi que leur capacité à acquérir des données dans n’importe quelles conditions atmosphériques de jour comme de nuit. Au cours des trois dernières décennies, plusieurs RSO polarimétriques ont été utilisés portant une variété de modes d’imagerie, tels que la polarisation unique, la polarisation double et également des modes dits pleinement polarimétriques. Grâce aux recherches récentes, d’autres modes alternatifs, tels que la polarisation hybride et compacte, ont été proposés pour les futures missions RSOs. Toutefois, un débat anime la communauté de la télédétection quant à l’utilité des modes alternatifs et quant au compromis entre la polarimétrie double et la polarimétrie totale. Cette thèse contribue à ce débat en analysant et comparant ces différents modes d’imagerie RSO dans une variété d’applications, avec un accent particulier sur la surveillance maritime (la détection des navires et de marées noires). Pour nos comparaisons, nous considérons un paramètre fondamental, appelé le degré de polarisation (DoP). Ce paramètre scalaire a été reconnu comme l’un des paramètres les plus pertinents pour caractériser les ondes électromagnétiques partiellement polarisées. A l’aide d’une analyse statistique détaillée sur les images polarimétriques RSO, nous proposons des estimateurs efficaces du DoP pour les systèmes d’imagerie cohérente et incohérente. Ainsi, nous étendons la notion de DoP aux différents modes d’imagerie polarimétrique hybride et compacte. Cette étude comparative réalisée dans différents contextes d’application dégage des propriétés permettant de guider le choix parmi les différents modes polarimétriques. Les expériences sont effectuées sur les données polarimétriques provenant du satellite Canadian RADARSAT-2 et le RSO aéroporté Américain AirSAR, couvrant divers types de terrains tels que l’urbain, la végétation et l’océan. Par ailleurs nous réalisons une étude détaillée sur les potentiels du DoP pour la détection et la reconnaissance des marées noires basée sur les acquisitions récentes d’UAVSAR, couvrant la catastrophe de Deepwater Horizon dans le golfe du Mexique. / Polarimetric Synthetic Aperture Radar (SAR) systems have become highly fruitful thanks to their wide area coverage and day and night all-weather capabilities. Several polarimetric SARs have been flown over the last few decades with a variety of polarimetric SAR imaging modes; traditional ones are linear singleand dual-pol modes. More sophisticated ones are full-pol modes. Other alternative modes, such as hybrid and compact dual-pol, have also been recently proposed for future SAR missions. The discussion is vivid across the remote sensing society about both the utility of such alternative modes, and also the trade-off between dual and full polarimetry. This thesis contributes to that discussion by analyzing and comparing different polarimetric SAR modes in a variety of geoscience applications, with a particular focus on maritime monitoring and surveillance. For our comparisons, we make use of a fundamental, physically related discriminator called the Degree of Polarization (DoP). This scalar parameter has been recognized as one of the most important parameters characterizing a partially polarized electromagnetic wave. Based on a detailed statistical analysis of polarimetric SAR images, we propose efficient estimators of the DoP for both coherent and in-coherent SAR systems. We extend the DoP concept to different hybrid and compact SAR modes and compare the achieved performance with different full-pol methods. We perform a detailed study of vessel detection and oil-spill recognition, based on linear and hybrid/compact dual-pol DoP, using recent data from the Deepwater Horizon oil-spill, acquired by the National Aeronautics and Space Administration (NASA)/Jet Propulsion Laboratory (JPL) Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR). Extensive experiments are also performed over various terrain types, such as urban, vegetation, and ocean, using the data acquired by the Canadian RADARSAT-2 and the NASA/JPL Airborne SAR (AirSAR) system.

In this work, we study the so-called combined models, generalized linear mixed models with extension to allow for overdispersion, in the context of genetics and breeding. Such flexible models accommodates cluster-induced correlation and overdispersion through two separate sets of random effects and contain as special cases the generalized linear mixed models (GLMM) on the one hand, and commonly known overdispersion models on the other. We use such models while obtaining heritability coefficients for non-Gaussian characters. Heritability is one of the many important concepts that are often quantified upon fitting a model to hierarchical data. It is often of importance in plant and animal breeding. Knowledge of this attribute is useful to quantify the magnitude of improvement in the population. For data where linear models can be used, this attribute is conveniently defined as a ratio of variance components. Matters are less simple for non-Gaussian outcomes. The focus is on time-to-event and count traits, where the Weibull-Gamma-Normal and Poisson-Gamma-Normal models are used. The resulting expressions are sufficiently simple and appealing, in particular in special cases, to be of practical value. The proposed methodologies are illustrated using data from animal and plant breeding. Furthermore, attention is given to the occurrence of negative estimates of variance components in the Poisson-Gamma-Normal model. The occurrence of negative variance components in linear mixed models (LMM) has received a certain amount of attention in the literature whereas almost no work has been done for GLMM. This phenomenon can be confusing at first sight because, by definition, variances themselves are non-negative quantities. However, this is a well understood phenomenon in the context of linear mixed modeling, where one will have to make a choice between a hierarchical and a marginal view. The variance components of the combined model for count outcomes are studied theoretically and the plant breeding study used as illustration underscores that this phenomenon can be common in applied research. We also call attention to the performance of different estimation methods, because not all available methods are capable of extending the parameter space of the variance components. Then, when there is a need for inference on such components and they are expected to be negative, the accuracy of the method is not the only characteristic to be considered. / Neste trabalho foram estudados os chamados modelos combinados, modelos lineares generalizados mistos com extensão para acomodar superdispersão, no contexto de genética e melhoramento. Esses modelos flexíveis acomodam correlação induzida por agrupamento e superdispersão por meio de dois conjuntos separados de efeitos aleatórios e contem como casos especiais os modelos lineares generalizados mistos (MLGM) e os modelos de superdispersão comumente conhecidos. Tais modelos são usados na obtenção do coeficiente de herdabilidade para caracteres não Gaussianos. Herdabilidade é um dos vários importantes conceitos que são frequentemente quantificados com o ajuste de um modelo a dados hierárquicos. Ela é usualmente importante no melhoramento vegetal e animal. Conhecer esse atributo é útil para quantificar a magnitude do ganho na população. Para dados em que modelos lineares podem ser usados, esse atributo é convenientemente definido como uma razão de componentes de variância. Os problemas são menos simples para respostas não Gaussianas. O foco aqui é em características do tipo tempo-até-evento e contagem, em que os modelosWeibull-Gama-Normal e Poisson-Gama-Normal são usados. As expressões resultantes são suficientemente simples e atrativas, em particular nos casos especiais, pelo valor prático. As metodologias propostas são ilustradas usando dados de melhoramento animal e vegetal. Além disso, a atenção é voltada à ocorrência de estimativas negativas de componentes de variância no modelo Poisson-Gama- Normal. A ocorrência de componentes de variância negativos em modelos lineares mistos (MLM) tem recebido certa atenção na literatura enquanto quase nenhum trabalho tem sido feito para MLGM. Esse fenômeno pode ser confuso a princípio porque, por definição, variâncias são quantidades não-negativas. Entretanto, este é um fenômeno bem compreendido no contexto de modelagem linear mista, em que a escolha deverá ser feita entre uma interpretação hierárquica ou marginal. Os componentes de variância do modelo combinado para respostas de contagem são estudados teoricamente e o estudo de melhoramento vegetal usado como ilustração confirma que esse fenômeno pode ser comum em pesquisas aplicadas. A atenção também é voltada ao desempenho de diferentes métodos de estimação, porque nem todos aqueles disponíveis são capazes de estender o espaço paramétrico dos componentes de variância. Então, quando há a necessidade de inferência de tais componentes e é esperado que eles sejam negativos, a acurácia do método de estimação não é a única característica a ser considerada.