We develop statistical methods for tackling two important problems in genetic association studies. First, we propose a Bayesian approach to overcome the winner's curse in genetic studies. Second, we consider a Bayesian latent variable model for analyzing longitudinal family data with pleiotropic phenotypes. Winner's curse in genetic association studies refers to the estimation bias of the reported odds ratios (OR) for an associated genetic variant from the initial discovery samples. It is a consequence of the sequential procedure in which the estimated effect of an associated genetic marker must first pass a stringent significance threshold. We propose a hierarchical Bayes method in which a spike-and-slab prior is used to account for the possibility that the significant test result may be due to chance. We examine the robustness of the method using different priors corresponding to different degrees of confidence in the testing results and propose a Bayesian model averaging procedure to combine estimates produced by different models. The Bayesian estimators yield smaller variance compared to the conditional likelihood estimator and outperform the latter in the low power studies. We investigate the performance of the method with simulations and applications to four real data examples. Pleiotropy occurs when a single genetic factor influences multiple quantitative or qualitative phenotypes, and it is present in many genetic studies of complex human traits. The longitudinal family studies combine the features of longitudinal studies in individuals and cross-sectional studies in families. Therefore, they provide more information about the genetic and environmental factors associated with the trait of interest. We propose a Bayesian latent variable modeling approach to model multiple phenotypes simultaneously in order to detect the pleiotropic effect and allow for longitudinal and/or family data. An efficient MCMC algorithm is developed to obtain the posterior samples by using hierarchical centering and parameter expansion techniques. We apply spike and slab prior methods to test whether the phenotypes are significantly associated with the latent disease status. We compute Bayes factors using path sampling and discuss their application in testing the significance of factor loadings and the indirect fixed effects. We examine the performance of our methods via extensive simulations and apply them to the blood pressure data from a genetic study of type 1 diabetes (T1D) complications.

Em um contexto de analise de dados de sobrevivência univariados ou multivariados, dados de tempos de falha caracterizam-se pela possibilidade de poderem ser censurados. Embora comum na pratica, a censura impede o uso de alguns procedimentos estatisticos covencionais o que vem motivando, em especial apos a publicacao do artigo de Cox (1972), o desenvolvimento de metodos estatisticos nessa area. Uma linha de estudo recente e a de que, em algumas situacoes, a variavel resposta esteja sendo inuenciada por variaveis latentes, variaveis estas que sao usadas, em um sentido estatistico, para descreverem efeitos geneticos ou ambientais compartilhados pelos indivduos ou, ainda, covariaveis nao consideradas no estudo. Nesse trabalho, enfase e dada aos modelos de sobrevivencia que consideram tempos de falha multivariados e variaveis latentes. Esses tempos aparecem quando, por exemplo, cada individuo em estudo esta sujeito a diversos eventos ou, quando existe um agrupamento natural ou artificial o qual induz dependencia entre os tempos dos individuos do mesmo grupo. Modelos com variaveis latentes em que tais tempos de falha ocorrem em intervalos de tempo, ou seja, em um contexto de censura intervalar sao especialmente considerados nesse trabalho. O modelo de fragilidade gama para dados de sobrevivencia com censura intervalar e proposto, nesse trabalho, como um criterio para a selecao de bovinos. Como uma alternativa para esta selecao, o modelo de curvas de crescimento com efeitos aleatorios e tambem considerado. Para a estimacao dos parametros envolvidos em ambos os modelos propostos, programas computacionais sao apresentados. Uma abordagem Bayesiana e considerada no processo de estimação sendo, o metodo de Markov chain Monte Carlo (MCMC) utilizado e as distribuicoes a posteriori obtidas, usando-se o amostrador de Gibbs. O modelo de fragilidade gama com censura intervalar e o de curvas de crescimento com efeitos aleatorios sao comparados por meio de um estudo de simulação. Para ilustrar ambos os modelos propostos, estudos com bovinos das racas Nelore e Canchim são utilizados. / In a context of univariate or multivariate survival data analysis, failure times data are characterized by the possibility to be censored. Although common in practice, censoring precludes the use of some conventional statistical procedures and it has been motivating, specially after the publication of the Cox's paper (1972), the development of statistical methods in this area. A recent topic of study is concerned with some situations where the response variable is in uenced by latent variables which are used in a statistical sense to describe genetic or environmental efects shared by individuals or also covariates not considered in the study. In this work emphasis is given to survival models which consider multivariate failure times and latent variables. Such times occur when, for instance, each individual under study is exposed to several events or when there is a natural or artificial clustering that causes dependence among times of those individuals at the same cluster. Models with latent variables where such failure times lie in intervals of time, i.e. in an interval censored context are specially considered in this work. The gamma frailty interval censored survival model is proposed in this work as a selection criterion for cattle. As an alternative selection criterion the growth curves model with random efects is also considered. To estimate the involved parameters in both proposed models, computational programs are presented. A Bayesian approach is considered in the estimation process so that the Markov chain Monte Carlo (MCMC) method is used and the posterior distributions are obtained using Gibbs sampling. The gamma frailty interval-censored survival model and the growth curves model with random efects are compared using a simulation study. To illustrate both proposed models studies with Nelore and Canchim cattle are used.

Cette thèse aborde la problématique de la classification non supervisée de données lorsque les caractéristiques des classes sont susceptibles d'évoluer au cours du temps. On parlera également, dans ce cas, de classification dynamique de données temporelles non stationnaires. Le cadre applicatif des travaux concerne le diagnostic par reconnaissance des formes de systèmes complexes dynamiques dont les classes de fonctionnement peuvent, suite à des phénomènes d'usures, des déréglages progressifs ou des contextes d'exploitation variables, évoluer au cours du temps. Un modèle probabiliste dynamique, fondé à la fois sur les mélanges de lois et sur les modèles dynamiques à espace d'état, a ainsi été proposé. Compte tenu de la structure complexe de ce modèle, une variante variationnelle de l'algorithme EM a été proposée pour l'apprentissage de ses paramètres. Dans la perspective du traitement rapide de flux de données, une version séquentielle de cet algorithme a également été développée, ainsi qu'une stratégie de choix dynamique du nombre de classes. Une série d'expérimentations menées sur des données simulées et des données réelles acquises sur le système d'aiguillage des trains a permis d'évaluer le potentiel des approches proposées / Nowadays, diagnosis and monitoring for predictive maintenance of railway components are important key subjects for both operators and manufacturers. They seek to anticipate upcoming maintenance actions, reduce maintenance costs and increase the availability of rail network. In order to maintain the components at a satisfactory level of operation, the implementation of reliable diagnostic strategy is required. In this thesis, we are interested in a main component of railway infrastructure, the railway switch; an important safety device whose failure could heavily impact the availability of the transportation system. The diagnosis of this system is therefore essential and can be done by exploiting sequential measurements acquired successively while the state of the system is evolving over time. These measurements consist of power consumption curves that are acquired during several switch operations. The shape of these curves is indicative of the operating state of the system. The aim is to track the temporal dynamic evolution of railway component state under different operating contexts by analyzing the specific data in order to detect and diagnose problems that may lead to functioning failure. This thesis tackles the problem of temporal data clustering within a broader context of developing innovative tools and decision-aid methods. We propose a new dynamic probabilistic approach within a temporal data clustering framework. This approach is based on both Gaussian mixture models and state-space models. The main challenge facing this work is the estimation of model parameters associated with this approach because of its complex structure. In order to meet this challenge, a variational approach has been developed. The results obtained on both synthetic and real data highlight the advantage of the proposed algorithms compared to other state of the art methods in terms of clustering and estimation accuracy

Neste trabalho apresentamos os mais importantes processos de estimação em algumas classes de modelos de resposta ao item (Dicotômicos e Policotômicos). Discutimos algumas propriedades desses métodos. Com o objetivo de comparar o desempenho dos métodos conduzimos simulações apropriadas. / In this work we show the most important estimation methods for some item response models (both dichotomous and polichotomous). We discuss some proprieties of these methods. To compare the characteristic of these methods we conducted appropriate simulations.

L'objectif de cette thèse est de concevoir des algorithmes qui permettent la détection robuste d’objets et d’événements dans des vidéos en s’appuyant sur une analyse conjointe de données audio et visuelle. Ceci est inspiré par la capacité remarquable des humains à intégrer les caractéristiques auditives et visuelles pour améliorer leur compréhension de scénarios bruités. À cette fin, nous nous appuyons sur deux types d'associations naturelles entre les modalités d'enregistrements audiovisuels (réalisés à l'aide d'un seul microphone et d'une seule caméra), à savoir la corrélation mouvement/audio et la co-occurrence apparence/audio. Dans le premier cas, nous utilisons la séparation de sources audio comme application principale et proposons deux nouvelles méthodes dans le cadre classique de la factorisation par matrices non négatives (NMF). L'idée centrale est d'utiliser la corrélation temporelle entre l'audio et le mouvement pour les objets / actions où le mouvement produisant le son est visible. La première méthode proposée met l'accent sur le couplage flexible entre les représentations audio et de mouvement capturant les variations temporelles, tandis que la seconde repose sur la régression intermodale. Nous avons séparé plusieurs mélanges complexes d'instruments à cordes en leurs sources constituantes en utilisant ces approches.Pour identifier et extraire de nombreux objets couramment rencontrés, nous exploitons la co-occurrence apparence/audio dans de grands ensembles de données. Ce mécanisme d'association complémentaire est particulièrement utile pour les objets où les corrélations basées sur le mouvement ne sont ni visibles ni disponibles. Le problème est traité dans un contexte faiblement supervisé dans lequel nous proposons un framework d’apprentissage de représentation pour la classification robuste des événements audiovisuels, la localisation des objets visuels, la détection des événements audio et la séparation de sources.Nous avons testé de manière approfondie les idées proposées sur des ensembles de données publics. Ces expériences permettent de faire un lien avec des phénomènes intuitifs et multimodaux que les humains utilisent dans leur processus de compréhension de scènes audiovisuelles. / The goal of this thesis is to design algorithms that enable robust detection of objectsand events in videos through joint audio-visual analysis. This is motivated by humans’remarkable ability to meaningfully integrate auditory and visual characteristics forperception in noisy scenarios. To this end, we identify two kinds of natural associationsbetween the modalities in recordings made using a single microphone and camera,namely motion-audio correlation and appearance-audio co-occurrence.For the former, we use audio source separation as the primary application andpropose two novel methods within the popular non-negative matrix factorizationframework. The central idea is to utilize the temporal correlation between audio andmotion for objects/actions where the sound-producing motion is visible. The firstproposed method focuses on soft coupling between audio and motion representationscapturing temporal variations, while the second is based on cross-modal regression.We segregate several challenging audio mixtures of string instruments into theirconstituent sources using these approaches.To identify and extract many commonly encountered objects, we leverageappearance–audio co-occurrence in large datasets. This complementary associationmechanism is particularly useful for objects where motion-based correlations are notvisible or available. The problem is dealt with in a weakly-supervised setting whereinwe design a representation learning framework for robust AV event classification,visual object localization, audio event detection and source separation.We extensively test the proposed ideas on publicly available datasets. The experimentsdemonstrate several intuitive multimodal phenomena that humans utilize on aregular basis for robust scene understanding.

This study investigates the role social information and personal dispositions play in the development of attitudes of unionized employees towards employee involvement programs. A theoretical model was developed in order to understand how social information and dispositions form union member attitudes towards employee involvement programs. This was designed from models of employee involvement and attitude formation. Data were collected from employees at electrical power generation facilities. Measures of organizational and union commitment, locus of control, participativeness, social information provided by the company, social information provided by the union, and employee involvement attitudes were gathered through a survey distributed at the facilities. General affect and satisfaction towards four types of employee involvement programs union members are most likely to encounter were measured. Specific hypotheses were developed in order to test and analyze parts of the theoretical model. While the results were at times contrary to the hypothesized relationships within the model, the data fit with the theorized model well enough to provide support for it. This model effectively demonstrated how employee involvement attitudes are formed from such data, and the relationships between the variables measured. / Ph. D.

A Monte Carlo simulation study on class enumeration with latent class regression models. / Latent class regression (LCR) is a statistical method used to identify qualitatively different groups or latent classes within a heterogeneous population and commonly used in the behavioural, health, and social sciences. Despite the vast applications, an agreed fit index to correctly determine the number of latent classes is hotly debated. To add, there are also conflicting views on whether covariates should or should not be included into the class enumeration process. We conduct a simulation study to determine the impact of covariates on the class enumeration accuracy as well as study the performance of several commonly used fit indices under different population models and modelling conditions. Our results indicate that of the eight fit indices considered, the aBIC and BLRT proved to be the best performing fit indices for class enumeration. Furthermore, we found that covariates should not be included into the enumeration procedure. Our results illustrate that an unconditional LCA model can enumerate equivalently as well as a conditional LCA model with its true covariate specification. Even with the presence of large covariate effects in the population, the unconditional model is capable of enumerating with high accuracy. As noted by Nylund and Gibson (2016), a misspecified covariate specification can easily lead to an overestimation of latent classes. Therefore, we recommend to perform class enumeration without covariates and determine a set of candidate latent class models with the aBIC. Once that is determined, the BLRT can be utilized on the set of candidate models and confirm whether results obtained by the BLRT match the results of the aBIC. By separating the enumeration procedure of the BLRT, it still allows one to use the BLRT but reduce the heavy computational burden that is associated with this fit index. Subsequent analysis can then be pursued accordingly after the number of latent classes is determined. / Thesis / Master of Science (MSc)

Thesis advisor: Henry I. Braun / Secondary analyses of international large-scale assessments (ILSA) commonly characterize relationships between variables of interest using correlations. However, the accuracy of correlation estimates is impaired by artefacts such as measurement error and clustering. Despite advancements in methodology, conventional correlation estimates or statistical models not addressing this problem are still commonly used when analyzing ILSA data. This dissertation examines the impact of both the clustered nature of the data and heterogeneous measurement error on the correlations reported between background data and proficiency scales across countries participating in ILSA. In this regard, the operating characteristics of competing modeling techniques are explored by means of applications to data from PISA 2012. Specifically, the estimates of correlations between math self-efficacy and math achievement across countries are the principal focus of this study. Sequentially employing four different statistical techniques, a step-wise model refinement approach is used. After each step, the changes in the within-country correlation estimates are examined in relation to (i) the heterogeneity of distributions, (ii) the amount of measurement error, (iii) the degree of clustering, and (iv) country-level math performance. The results show that correlation estimates gathered from two-dimensional IRT models are more similar across countries in comparison to conventional and multilevel linear modeling estimates. The strength of the relationship between math proficiency and math self-efficacy is moderated by country mean math proficiency and this was found to be consistent across all four models even when measurement error and clustering were taken into account. Multilevel multidimensional mixture IRT modeling results support the hypothesis that low-performing groups within countries have a lower correlation between math self-efficacy and math proficiency. A weaker association between math self-efficacy and math proficiency in lower achieving groups is consistently seen across countries. A multilevel mixture IRT modeling approach sheds light on how this pattern emerges from greater randomness in the responses of lower performing groups. The findings from this study demonstrate that advanced modeling techniques not only are more appropriate given the characteristics of the data, but also provide greater insight about the patterns of relationships across countries. / Thesis (PhD) — Boston College, 2021. / Submitted to: Boston College. Lynch School of Education. / Discipline: Educational Research, Measurement and Evaluation.

No description available.

<p>Accommodating changing consumer tastes, nutritional targets, competitive pressures and government regulations is an ongoing task in the food industry. Product development projects tend to have competing goals and more potential solutions than can be examined efficiently. However, existing databases or spreadsheets containing formulas, ingredient properties, and product characteristics can be exploited using latent variable methods to confront difficult formulation issues. Using these methods, a product developer can target specific final product properties and systematically determine new recipes that will best meet the development objectives.</p> <p>Latent variable methods in reformulation are demonstrated for a product line of frozen muffin batters used in the food service industry. A particular attribute is to be minimized while maintaining the taste, texture, and appearance of the original products, but the minimization is difficult because the attribute in question is not well understood. Initially, existing data is used to develop a partial least squares (PLS) model, which identifies areas for further testing. Design of experiments (DOE) in the latent variable space generates new data that is used to augment the model. An optimization algorithm makes use of the updated model to produce recipes for four different products, and a significant reduction of the target attribute is achieved in all cases.</p> <p>Latent variable methods are also applied to a difficult classification problem in oat milling. Process monitoring involves manually classifying and counting the oats and hulls in the product streams of groats; a task that is time-consuming and therefore infrequent. A solution based on near infrared (NIR) imaging and PLS-discriminant analysis (PLS-DA) is investigated and found to be feasible. The PLS-DA model, built using mixed-cultivar samples, effectively separates the oats and groats into two classes. The model is validated using samples of three pure cultivars with varying moistures and growing conditions.</p> / Master of Applied Science (MASc)