Littérature scientifique sur le sujet « Dynamic structural equation models »

The thesis focuses on estimation of dynamic structural equation models in which some or all variables might be unobservable (latent) or measured with error. Moreover, we consider the situation where latent variables can be measured with multiple observable indicators and where lagged values of latent variables might be included in the model. This situation leads to a dynamic structural equation model (DSEM), which can be viewed as dynamic generalisation of the structural equation model (SEM). Taking the mismeasurement problem into account aims at reducing or eliminating the errors-in-variables bias and hence at minimising the chance of obtaining incorrect coefficient estimates. Furthermore, such methods can be used to improve measurement of latent variables and to obtain more accurate forecasts. The thesis aims to make a contribution to the literature in four areas. Firstly, we propose a unifying theoretical framework for the analysis of dynamic structural equation models. Secondly, we provide analytical results for both panel and time series DSEM models along with the software implementation suggestions. Thirdly, we propose non-parametric estimation methods that can also be used for obtaining starting values in maximum likelihood estimation. Finally, we illustrate these methods on several real data examples demonstrating the capabilities of the currently available software as well as importance of good starting values.

Structural equation modeling (SEM) is often used to investigate effective connectivity in functional neuroimaging studies. Modeling effective connectivity refers to an approach in which a number of specific brain regions, called regions of interest (ROIs), are selected according to some prior knowledge about the regions, and directional (causal) relationships between them are hypothesized and tested. Existing methods for SEM, however, have serious limitations in terms of their computational capacity and the range of models that can be specified. To alleviate these difficulties, I propose a new method of SEM for analysis of effective connectivity, called Dynamic GSCA (Generalized Structured Component Analysis). This method is a component-based method that combines the original GSCA and a multivariate autoregressive model to account for the dynamic nature of data taken over time. Dynamic GSCA can accommodate more elaborate structural models that describe relationships among ROIs and is less prone to computational difficulties, such as improper solutions and the lack of model identification, than the conventional methods of SEM. To illustrate the use of the proposed method, results of empirical studies based on synthetic and real data are reported. Further extensions of Dynamic GSCA are also discussed, including higher order components, multi-sample comparison, multilevel analysis, and latent interactions.<br>La Modélisation par Équations Structurelles (MES) est souvent utilisée dans les études d'imagerie cérébrales fonctionnelles afin d'investiguer la connectivité effective. La modélisation de connectivité effective est une approche dans laquelle certaines régions cérébrales, appelées régions d'intérêts (RIs), sont spécifiquement sélectionnées à partir de connaissances établies sur ces régions, et des hypothèses sur les possibles liens directionnels (causals) entre les RIs sont formulées et testées. Par contre, les méthodes de MES existantes sont sérieusement limitées par leur capacité computationelle et le nombre et l'étendue des modèles qui peuvent être spécifiés. Afin d'adresser ces difficultés, je propose ici une nouvelle méthode de MES afin d'analyser la connectivité effective, appelée Analyse en Composantes Structurée Généralisée (ACSG) Dynamique. Cette méthode est une méthode basée sur les composantes, combinant la version originale des ACSGs et un modèle auto-régresseur multi-variable afin de tenir compte de la nature dynamique des données recueillies à différent temps. Les ACSG Dynamiques peuvent accommoder des modèles structurels plus complexes pour décrire les relations entre les RIs. De plus, comparé aux méthodes traditionnelles de MES, les ACSG Dynamiques sont moins susceptible de succomber aux difficultés computationelles, comme les solutions inappropriées et l'échec d'identification de modèle. Afin d'illustrer l'utilisation de la méthode proposée, des résultats d'études empiriques basées sur des données synthétiques et réelles sont présentées. Des extensions possibles des ACSG Dynamiques sont aussi discutées, incluant des composantes de plus haut niveau, la comparaison de plusieurs échantillons, l'analyse multi-niveau, et les interactions latentes.

Effective connectivity in functional neuroimaging studies is defined as the time dependent causal influence that a certain brain region of interest (ROI) exerts on another. Structural equation modeling (SEM) is regularly employed to analyze effective connectivity. In recent years, various SEM methods have been proposed to model effective connectivity. However, there has been little attempt to develop SEM methods for analyzing common patterns of effective connectivity across subjects despite the prevalence of multiple-subject research in effective connectivity. This dissertation proposes a method that fills this gap. This method is called dynamic generalized (multiple-set) structured canonical correlation analysis (dynamic GCANO). It combines generalized (multiple-set) canonical correlation analysis (GCANO) with a multivariate autoregressive time series model in a unified framework. This dissertation begins with a brief review of existing SEM techniques, and points out their limitations in analyzing multiple-subject effective connectivity data, which serves as a motivation to develop dynamic GCANO. The technical underpinnings of the proposed method are then stated, including specifications of a modeling framework and an optimization criterion for parameter estimation, which is minimized by an alternating least squares algorithm. The effectiveness of dynamic GCANO is demonstrated by analyzing both synthetic and real data sets. The former reveals reasonably good parameter recoveries by the proposed method, while the latter shows the usefulness of the method in empirical research. Several features of dynamic GCANO are highlighted through these examples. The dissertation concludes with possible extensions of the proposed method.<br>Suivant les méthodes d'imagerie fonctionnelle cérébrale, une connectivité efficace est définie comme influence dépendant de causalité temporelle qu'une certaine région d'intérêt du cerveau (ROI) exerce sur une autre. La modélisation par équation structurelle (SEM) est régulièrement utilisée pour analyser la connectivité efficace. Ces dernières années, diverses méthodes de SEM ont été proposées pour la modélisation de la connectivité efficace. Cependant, il y a eu peu de tentative pour développer des méthodes de SEM pour analyser les modèles communs de connectivité efficace sur-sujets, malgré la prédominance de recherche sur des sujets multiples pour analyser la connectivité efficace. Cette thèse propose une méthode qui comble cette lacune. Cette méthode est appelée dynamique généralisée (multiples ensemble) structuré l'analyse de corrélation canonique (dynamique GCANO). Elle combine généralisée (multiples ensemble) structuré l'analyse de corrélation canonique (GCANO) avec multivariée des séries chronologiques autorégressif dans un cadre unifié. Cette thèse commence par un bref sommaire sur les techniques existantes de SEM et souligne leurs limites pour analyser les données de plusieurs sous réserve pour la connectivité efficace, ce qui a mené à développer la dynamique GCANO. Les techniques de base de la méthode proposée sont ensuite énumérées, y compris les spécifications du cadre de modélisation et un critère d'optimisation pour l'estimation de paramètres, qui est réduit par alternant algorithme des moindres carrés. L'efficacité du dynamique GCANO est démontrée par l'analyse des ensembles de données synthétiques et réels. Les données synthétiques montrent une récupération raisonnable de paramètre par la méthode proposée, alors que les données réelles montrent l'utilité de la méthode dans les recherches empiriques. Plusieurs fonctionnalités du dynamique sont mises en évidence par le biais de ces exemples. En conclusion, la thèse propose des extensions possibles de la méthode proposée.

The present study tested the interactive model of affect, motivation, and engagement (Linnenbrink, 2007) in mathematics education with a nationally representative sample. Self-efficacy, self-concept, and anxiety were indicators of pleasant and unpleasant affect. Intrinsic and extrinsic motivation were indicators of mastery and performance approach. Persistence and cognitive activation were indicators of behavioral and cognitive engagement. The 2012 Programme for International Student Assessment (PISA) supplied a sample of 4,978 students from the United States for structural equation modeling. The results indicated that PISA data overall supported the interactive model. Specifically, PISA data completely supported the specification of the relationship between motivation and affect, largely supported the specification of the relationship between affect and engagement, but failed to support the specification of the relationship between motivation and engagement. Finally, PISA data largely supported the specification of the mediation effects of affect on the relationship between motivation and engagement.

Title from PDF of title page (University of Missouri--Columbia, viewed on Feb 26, 2010). The entire thesis text is included in the research.pdf file; the official abstract appears in the short.pdf file; a non-technical public abstract appears in the public.pdf file. Dissertation advisor: Dr. Scott Brown. Includes bibliographical references.

As organizações estão expostas a uma quantidade e variabilidade cada vez mais crescente de informações. A capacidade de se antecipar às tendências e se adequar ao ambiente passa a ser, além de fonte de vantagem competitiva, fator necessário para a sobrevivência. Nessa realidade, organizações frequentemente apresentam dificuldades de leitura de seu ambiente e adaptação ao meio. Acredita-se na necessidade de desenvolvimento de uma capacidade interna da organização para que o monitoramento do ambiente se estabeleça, bem como análise de oportunidades, planejamento de ações de melhoria e reconfiguração da organização. Este estudo buscou verificar a relação da Inteligência - enquanto processo de monitoramento do ambiente - como uma Capacidades Dinâmica, e de como esta pode contribuir com a geração de vantagem competitiva. Criou-se um modelo de pesquisa, utilizando-se dos modelos de Rios (2010) e Teece (2014), relacionando os conceitos de Inteligência e Capacidade Dinâmicas, e estas com a Vantagem Competitiva. Por meio de um questionário, realizou-se uma Pesquisa Survey, onde coletaram-se respostas de funcionários e/ou sócio de empresas brasileiras, independente de porte ou segmento. Para análise, utilizou-se da Modelagem de Equações Estruturais, e foi possível demonstrar que a Inteligência influencia positivamente nas Capacidades Dinâmicas do sub-grupo Transforming, na Estratégia e na Vantagem Competitiva. Dessa forma o modelo desenvolvido, tendo apresentado boa confiabilidade e aderência, pode também ser validado.<br>Organizations are exposed to an increasing amount and variability of information. The ability to anticipate trends and adapt to the environment becomes, besides a source of competitive advantage, a necessary factor for survival. In this reality, organizations frequently present difficulties in reading their environment and adapting to them. We believe in the need to develop an internal capacity of the organization for the monitoring of the environment to be established, as well as analysis of opportunities, planning of actions of improvement and reconfiguration of the organization. This study sought to verify the relationship of Intelligence - as a process of monitoring the environment - as a Dynamic Capabilities, and how this can contribute to the generation of competitive advantage. A research model was created, using the models of Rios (2010) and Teece (2014), relating the concepts of Dynamic Intelligence and Capacity, and these with the Competitive Advantage. By means of a questionnaire, a Survey Research was conducted, where responses were collected from employees and / or partners of Brazilian companies, regardless of size or segment. For the analysis, it was used the Modeling of Structural Equations, and it was possible to demonstrate that the Intelligence influences positively in the Dynamic Capacities of the Transforming subgroup, in the Strategy and the Competitive Advantage. In this way the developed model, having presented good reliability and adhesion, can also be validated.

Causal models have been used in different areas of knowledge in order to comprehend the causal associations between variables. Over the past decades, the amount of studies using these models have been growing a lot, especially those related to biological systems where studying and learning causal relationships among traits are essential for predicting the consequences of interventions in such system. Graph analysis (GA) and structural equation modeling (SEM) are tools used to explore such associations. While GA allows searching causal structures that express qualitatively how variables are causally connected, fitting SEM with a known causal structure allows to infer the magnitude of causal effects. Also SEM can be viewed as multiple regression models in which response variables can be explanatory variables for others. In quantitative genetics studies, SEM aimed to study the direct and indirect genetic effects associated to individuals through information related to them, beyond the observed characteristics, such as the kinship relations. In those studies typically the assumptions of linear relationships among traits are made. However, in some scenarios, nonlinear relationships can be observed, which make unsuitable the mentioned assumptions. To overcome this limitation, this paper proposes to use a mixed effects polynomial structural equation model, second or superior degree, to model those nonlinear relationships. Two studies were developed, a simulation and an application to real data. The first study involved simulation of 50 data sets, with a fully recursive causal structure involving three characteristics in which linear and nonlinear causal relations between them were allowed. The second study involved the analysis of traits related to dairy cows of the Holstein breed. Phenotypic relationships between traits were calving difficulty, gestation length and also the proportion of perionatal death. We compare the model of multiple traits and polynomials structural equations models, under different polynomials degrees in order to assess the benefits of the SEM polynomial of second or higher degree. For some situations the inappropriate assumption of linearity results in poor predictions of the direct, indirect and total of the genetic variances and covariance, either overestimating, underestimating, or even assign opposite signs to covariances. Therefore, we conclude that the inclusion of a polynomial degree increases the SEM expressive power.<br>Modelos causais têm sido muitos utilizados em estudos em diferentes áreas de conhecimento, a fim de compreender as associações ou relações causais entre variáveis. Durante as últimas décadas, o uso desses modelos têm crescido muito, especialmente estudos relacionados à sistemas biológicos, uma vez que compreender as relações entre características são essenciais para prever quais são as consequências de intervenções em tais sistemas. Análise do grafo (AG) e os modelos de equações estruturais (MEE) são utilizados como ferramentas para explorar essas relações. Enquanto AG nos permite buscar por estruturas causais, que representam qualitativamente como as variáveis são causalmente conectadas, ajustando o MEE com uma estrutura causal conhecida nos permite inferir a magnitude dos efeitos causais. Os MEE também podem ser vistos como modelos de regressão múltipla em que uma variável resposta pode ser vista como explanatória para uma outra característica. Estudos utilizando MEE em genética quantitativa visam estudar os efeitos genéticos diretos e indiretos associados aos indivíduos por meio de informações realcionadas aos indivíduas, além das característcas observadas, como por exemplo o parentesco entre eles. Neste contexto, é tipicamente adotada a suposição que as características observadas são relacionadas linearmente. No entanto, para alguns cenários, relações não lineares são observadas, o que torna as suposições mencionadas inadequadas. Para superar essa limitação, este trabalho propõe o uso de modelos de equações estruturais de efeitos polinomiais mistos, de segundo grau ou seperior, para modelar relações não lineares. Neste trabalho foram desenvolvidos dois estudos, um de simulação e uma aplicação a dados reais. O primeiro estudo envolveu a simulação de 50 conjuntos de dados, com uma estrutura causal completamente recursiva, envolvendo 3 características, em que foram permitidas relações causais lineares e não lineares entre as mesmas. O segundo estudo envolveu a análise de características relacionadas ao gado leiteiro da raça Holandesa, foram utilizadas relações entre os seguintes fenótipos: dificuldade de parto, duração da gestação e a proporção de morte perionatal. Nós comparamos o modelo misto de múltiplas características com os modelos de equações estruturais polinomiais, com diferentes graus polinomiais, a fim de verificar os benefícios do MEE polinomial de segundo grau ou superior. Para algumas situações a suposição inapropriada de linearidade resulta em previsões pobres das variâncias e covariâncias genéticas diretas, indiretas e totais, seja por superestimar, subestimar, ou mesmo atribuir sinais opostos as covariâncias. Portanto, verificamos que a inclusão de um grau de polinômio aumenta o poder de expressão do MEE.

In structural equation models with latent variables, maximum likelihood (ML) estimation is currently the most prevailing estimation method. However, the ML method fails to provide accurate solutions in a number of situations including those involving small sample sizes, nonnormality, and model misspecification. To over come these difficulties, regularized extensions of two-stage least squares estimation are proposed that incorporate a ridge type of regularization in the estimation of parameters. Two simulation studies and two empirical applications demonstrate that the proposed method is a promising alternative to both the maximum likelihood and non-regularized two-stage least squares estimation methods. An optimal value of the regularization parameter is found by the K-fold cross validation technique. A nonparametric bootstrap method is used to evaluate the stability of solutions. A goodness-of-fit measure is used for assessing the overall fit.<br>Dans les modèles d'équations structurales avec des variables latentes, l'estimation demaximum devraisemblance est la méthode d'estimation la plus utilisée. Par contre, la méthode de maximum devraisemblance souvent ne réussit pas á fournir des solutions exactes, par exemple lorsque les échantillons sont petits, les données ne sont pas normale, ou lorsque le modèle est mal specifié. L'estimation des moindres carrés á deux-phases est asymptotiquement sans distribution et robuste contre mauvaises spécifications, mais elle manque de robustesse quand les chantillons sont petits. Afin de surmonter les trois difficultés mentionnés ci-dessus et d'obtenir une estimation plus exacte, des extensions régularisées des moindres carrés á deux phases sont proposé á qui incorporent directement un type de régularisation dans les modèles d'équations structurales avec des variables latentes. Deux études de simulation et deux applications empiriques démontrent que la méthode propose est une alternative prometteuse aux méthodes de maximum vraisemblance et de l'estimation des moindres carrés á deux-phases. Un paramètre de régularisation valeur optimale a été trouvé par la technique de validation croisé d'ordre K. Une méthode non-paramétrique Bootstrap est utilisée afin d'évaluer la stabilité des solutions. Une mesure d'adéquation est utilisée pour estimer l'adéquation globale.