Often predictor variables in regression models are measured with errors. This is known as an errors-in-variables (EIV) problem. The statistical analysis of the data ignoring the EIV is called naive analysis. As a result, the variance of the errors is underestimated. This affects any statistical inference that may subsequently be made about the model parameter estimates or the response prediction. In some cases (e.g. quadratic polynomial models) the parameter estimates and the model prediction is biased. The errors can occur in different ways. These errors are mainly classified into classical (i.e. occur in observational studies) or Berkson type (i.e. occur in designed experiments). This thesis addresses the problem of the Berkson EIV and their effect on the statistical analysis of data fitted using linear and nonlinear models. In particular, the case when the errors are dependent and have heterogeneous variance is studied. Both analytical and empirical tools have been used to develop new approaches for dealing with this type of errors. Two different scenarios are considered: mixture experiments where the model to be estimated is linear in the parameters and the EIV are correlated; and bioassay dose-response studies where the model to be estimated is nonlinear. EIV following Gaussian distribution, as well as the much less investigated non-Gaussian distribution are examined. When the errors occur in mixture experiments both analytical and empirical results showed that the naive analysis produces biased and inefficient estimators for the model parameters. The magnitude of the bias depends on the variances of the EIV for the mixture components, the model and its parameters. First and second Scheffé polynomials are used to fit the response. To adjust for the EIV, four different approaches of corrections are proposed. The statistical properties of the estimators are investigated, and compared with the naive analysis estimators. Analytical and empirical weighted regression calibration methods are found to give the most accurate and efficient results. The approaches require the error variance to be known prior to the analysis. The robustness of the adjusted approaches for misspecified variance was also examined. Different error scenarios of EIV in the settings of concentrations in bioassay dose-response studies are studied (i.e. dependent and independent errors). The scenarios are motivated by real-life examples. Comparisons between the effects of the errors are illustrated using the 4-prameter Hill model. The results show that when the errors are non-Gaussian, the nonlinear least squares approach produces biased and inefficient estimators. An extension of the well-known simulation-extrapolation (SIMEX) method is developed for the case when the EIV lead to biased model parameters estimators, and is called Berkson simulation-extrapolation (BSIMEX). BSIMEX requires the error variance to be known. The robustness of the adjusted approach for misspecified variance is examined. Moreover, it is shown that BSIMEX performs better than the regression calibration methods when the EIV are dependent, while the regression calibration methods are preferable when the EIV are independent.

Modelos com erros de medição têm recebido a atenção de vários pesquisadores das mais diversas áreas de conhecimento. O principal objetivo desta dissertação consiste no estudo de um modelo funcional com erros de medição heteroscedásticos na presença de réplicas das observações. O modelo proposto estende resultados encontrados na literatura na medida em que as réplicas são parte do modelo, ao contrário de serem utilizadas para estimação das variâncias, doravante tratadas como conhecidas. Alguns procedimentos de estimação tais como o método de máxima verossimilhança, o método dos momentos e o método de extrapolação da simulação (SIMEX) na versão empírica são apresentados. Além disso, propõe-se o teste da razão de verossimilhanças e o teste de Wald com o objetivo de testar algumas hipóteses de interesse relacionadas aos parâmetros do modelo adotado. O comportamento dos estimadores de alguns parâmetros e das estatísticas propostas (resultados assintóticos) são analisados por meio de um estudo de simulação de Monte Carlo, utilizando-se diferentes números de réplicas. Por fim, a proposta é exemplificada com um conjunto de dados reais. Toda parte computacional foi desenvolvida em linguagem R (R Development Core Team, 2011) / Measurement error models have received the attention of many researchers of several areas of knowledge. The aim of this dissertation is to study a functional heteroscedastic measurement errors model with replicated observations. The proposed model extends results from the literature in that replicas are part of the model, as opposed to being used for estimation of the variances, now treated as known. Some estimation procedures such as maximum likelihood method, the method of moments and the empirical simulation-extrapolation method (SIMEX) are presented. Moreover, it is proposed the likelihood ratio test and Wald test in order to test hypotheses of interest related to the model parameters used. The behavior of the estimators of some parameters and statistics proposed (asymptotic results) are analyzed through Monte Carlo simulation study using different numbers of replicas. Finally, the proposal is illustrated with a real data set. The computational part was developed in R language (R Development Core Team, 2011)

No description available.

Para analisar características de interesse a respeito de um conjunto de dados reais da área de Odontologia apresentado em Hadgu & Koch (1999), ajustaremos um modelo de regressão linear multivariado com erros nas variáveis com intercepto nulo. Este conjunto de dados é caracterizado por medições de placa bacteriana em três grupos de voluntários, antes e após utilizar dois líquidos de bochecho experimentais e um líquido de bochecho controle, com medições (sujeitas a erros de medição) no início do estudo, após três e seis meses de utilização dos líquidos. Neste caso, uma possível estrutura de dependência entre as medições feitas em um mesmo indivíduo deve ser incorporada ao modelo e, além disto, temos duas variáveis resposta para cada indivíduo. Após a apresentação do modelo estatístico, iremos obter estimativas de máxima verossimilhança dos parâmetros utilizando o algoritmo iterativo EM e testaremos as hipóteses de interesse utilizando testes assintóticos de Wald, razão de verossimilhanças e score. Como neste caso não existe um teste ótimo, faremos um estudo de simulação para verificar o comportamento das três estatísticas de teste em relação a diferentes tamanhos amostrais e diferentes valores de parâmetros. Finalmente, faremos um estudo de diagnóstico buscando identificar possíveis pontos influentes no modelo, considerando o enfoque de influência local proposto por Cook (1986) e a medida de curvatura normal conformal desenvolvida por Poon & Poon (1999). / To analyze some characteristics of interest in a real odontological data set presented in Hadgu & Koch (1999), we propose the use of a multivariate null intercept errors-in-variables regression model. This data set is composed by measurements of dental plaque index (with measurement errors), which were measured in volunteers who were randomized to two experimental mouth rinses (A and B) or a control mouth rinse. The measurements were taken in each individual, before and after the use of the respective mouth rinses, in the beginning of the study, after three months from the baseline and after six months from the baseline. In this case, a possible structure of dependency between the measurements taken within the same individual must be incorporated in the model. After presenting the statistical model, we obtain the maximum likelihood estimates of the parameters using the numerical algorithm EM, and we test the hypotheses of interest considering asymptotic tests (Wald, likelihood ratio and score). Also, a simulation study to verify the behavior of these three test statistics is presented, considering diferent sample sizes and diferent values for the parameters. Finally, we make a diagnostic study to identify possible influential observations in the model, considering the local influence approach proposed by Cook (1986) and the conformal normal curvature proposed by Poon & Poon (1999).

A linear regression model, where covariates and a response are subject to errors, is considered in this thesis. For so-called errors-in-variables (EIV) model, suitable error structures are proposed, various unknown parameter estimation techniques are performed, and recent algebraic and statistical results are summarized. An extension of the total least squares (TLS) estimate in the EIV model-the EIV estimate-is invented. Its invariant (with respect to scale) and equivariant (with respect to the covariates' rotation, to the change of covariates direction, and to the interchange of covariates) properties are derived. Moreover, it is shown that the EIV estimate coincides with any unitarily invariant penalizing solution to the EIV problem. It is demonstrated that the asymptotic normality of the EIV estimate is computationally useless for a construction of confidence intervals or hypothesis testing. A proper bootstrap procedure is constructed to overcome such an issue. The validity of the bootstrap technique is proved. A simulation study and a real data example assure of its appropriateness. Strong and uniformly strong mixing errors are taken into account instead of the independent ones. For such a case, the strong consistency and the asymptotic normality of the EIV estimate are shown. Despite of that, their...

We investigate possible usage of Errors-in-Variables estimator (EIV), when esti- mating structural equations models (SEM). Structural equations modelling pro- vides framework for analysing complex relations among set of random variables where for example the response variable in one equation plays role of the predic- tor in another equation. First an overview of SEM and some common covariance based estimators is provided. Special case of linear regression model is investi- gated, showing that the covariance based estimators yield the same results as ordinary least squares. A compact review of EIV models follows, Errors-in-Variables models are re- gression models where not only response but also predictors are assumed to be measured with an error. Main contribution of this paper then lies in defining modifications of the EIV estimator to fit in the SEM framework. General opti- mization problem to estimate the parameters of structural equations model with errors-in-variables si postulated. Several modifications of two stage least squares are also proposed for future research. Equation-wise Errors-in-Variables estimator is proposed to estimate the coeffi- cients of structural equations model. The coefficients of every structural equation are estimated separately using EIV estimator. Some theoretical conditions...

Cette thèse traite du traitement adaptatif spatio-temporel dans le domaine radar. Pour augmenter les performances en détection, cette approche consiste à maximiser le rapport entre la puissance de la cible et celle des interférences, à savoir le bruit thermique et le fouillis. De nombreuses variantes de cet algorithme existent, une d’entre elles est fondée sur une modélisation autorégressive vectorielle des interférences. Sa principale difficulté réside dans l’estimation des matrices autorégressives à partir des données d’entrainement ; ce point constitue l’axe de notre travail de recherche. En particulier, notre contribution porte sur deux aspects. D’une part, dans le cas où l’on suppose que le bruit thermique est négligeable devant le fouillis non gaussien, les matrices autorégressives sont estimées en utilisant la méthode du point fixe. Ainsi, l’algorithme est robuste à la distribution non gaussienne du fouillis.D’autre part, nous proposons une nouvelle modélisation des interférences différenciant le bruit thermique et le fouillis : le fouillis est considéré comme un processus autorégressif vectoriel, gaussien et perturbé par le bruit blanc thermique. Ainsi, de nouvelles techniques d'estimation des matrices autorégressives sont proposées. La première est une estimation aveugle par bloc reposant sur la technique à erreurs dans les variables. Ainsi, l’estimation des matrices autorégressives reste robuste pour un rapport faible entre la puissance de la cible et celle du fouillis (< 5 dB). Ensuite, des méthodes récursives ont été développées. Elles sont fondées sur des approches du type Kalman : filtrage de Kalman étendu et filtrage par sigma point (UKF et CDKF), ainsi que sur le filtre H∞.Une étude comparative sur des données synthétiques et réelles, avec un fouillis gaussien ou non gaussien, est menée pour révéler la pertinence des différents estimateurs en terme de probabilité de détection. / This dissertation deals with space-time adaptive processing in the radar’s field. To improve the detection’s performances, this approach consists in maximizing the ratio between the target’s power and the interference’s one, i.e. the thermal noise and the clutter. Several variants of its algorithm exist, one of them is based on multichannel autoregressive modelling of interferences. Its main problem lies in the estimation of autoregressive matrices with training data and guides our research’s work. Especially, our contribution is twofold.On the one hand, when thermal noise is considered negligible, autoregressive matrices are estimated with fixed point method. Thus, the algorithm is robust against non-gaussian clutter.On the other hand, a new modelling of interferences is proposed. The clutter and thermal noise are separated : the clutter is considered as a multichannel autoregressive process which is Gaussian and disturbed by the white thermal noise. Thus, new estimation’s algorithms are developed. The first one is a blind estimation based on errors in variable methods. Then, recursive approaches are proposed and used extension of Kalman filter : the extended Kalman filter and the Sigma Point Kalman filter (UKF and CDKF), and the H∞ filter. A comparative study on synthetic and real data with Gausian and non Gaussian clutter is carried out to show the relevance of the different algorithms about detection’s probability.

Para analisar características de interesse a respeito de um conjunto de dados reais da área de Odontologia apresentado em Hadgu & Koch (1999), ajustaremos um modelo de regressão linear multivariado com erros nas variáveis com intercepto nulo. Este conjunto de dados é caracterizado por medições de placa bacteriana em três grupos de voluntários, antes e após utilizar dois líquidos de bochecho experimentais e um líquido de bochecho controle, com medições (sujeitas a erros de medição) no início do estudo, após três e seis meses de utilização dos líquidos. Neste caso, uma possível estrutura de dependência entre as medições feitas em um mesmo indivíduo deve ser incorporada ao modelo e, além disto, temos duas variáveis resposta para cada indivíduo. Após a apresentação do modelo estatístico, iremos obter estimativas de máxima verossimilhança dos parâmetros utilizando o algoritmo iterativo EM e testaremos as hipóteses de interesse utilizando testes assintóticos de Wald, razão de verossimilhanças e score. Como neste caso não existe um teste ótimo, faremos um estudo de simulação para verificar o comportamento das três estatísticas de teste em relação a diferentes tamanhos amostrais e diferentes valores de parâmetros. Finalmente, faremos um estudo de diagnóstico buscando identificar possíveis pontos influentes no modelo, considerando o enfoque de influência local proposto por Cook (1986) e a medida de curvatura normal conformal desenvolvida por Poon & Poon (1999). / To analyze some characteristics of interest in a real odontological data set presented in Hadgu & Koch (1999), we propose the use of a multivariate null intercept errors-in-variables regression model. This data set is composed by measurements of dental plaque index (with measurement errors), which were measured in volunteers who were randomized to two experimental mouth rinses (A and B) or a control mouth rinse. The measurements were taken in each individual, before and after the use of the respective mouth rinses, in the beginning of the study, after three months from the baseline and after six months from the baseline. In this case, a possible structure of dependency between the measurements taken within the same individual must be incorporated in the model. After presenting the statistical model, we obtain the maximum likelihood estimates of the parameters using the numerical algorithm EM, and we test the hypotheses of interest considering asymptotic tests (Wald, likelihood ratio and score). Also, a simulation study to verify the behavior of these three test statistics is presented, considering diferent sample sizes and diferent values for the parameters. Finally, we make a diagnostic study to identify possible influential observations in the model, considering the local influence approach proposed by Cook (1986) and the conformal normal curvature proposed by Poon & Poon (1999).

Indiana University-Purdue University Indianapolis (IUPUI) / Useful medical indices pose important roles in predicting medical outcomes. Medical indices, such as the well-known Body Mass Index (BMI), Charleson Comorbidity Index, etc., have been used extensively in research and clinical practice, for the quantification of risks in individual patients. However, the development of these indices is challenged; and primarily based on heuristic arguments. Statistically, most medical indices can be expressed as a function of a linear combination of individual variables and fitted by single-index model. Single-index model represents a way to retain latent nonlinear features of the data without the usual complications that come with increased dimensionality. In my dissertation, I propose a single-index model approach to analytically derive indices from observed data; the resulted index inherently correlates with specific health outcomes of interest. The first part of this dissertation discusses the derivation of an index function for the prediction of one outcome using longitudinal data. A cubic-spline estimation scheme for partially linear single-index mixed effect model is proposed to incorporate the within-subject correlations among outcome measures contributed by the same subject. A recursive algorithm based on the optimization of penalized least square estimation equation is derived and is shown to work well in both simulated data and derivation of a new body mass measure for the assessment of hypertension risk in children. The second part of this dissertation extends the single-index model to a multivariate setting. Specifically, a multivariate version of single-index model for longitudinal data is presented. An important feature of the proposed model is the accommodation of both correlations among multivariate outcomes and among the repeated measurements from the same subject via random effects that link the outcomes in a unified modeling structure. A new body mass index measure that simultaneously predicts systolic and diastolic blood pressure in children is illustrated. The final part of this dissertation shows existence, root-n strong consistency and asymptotic normality of the estimators in multivariate single-index model under suitable conditions. These asymptotic results are assessed in finite sample simulation and permit joint inference for all parameters.

Cette thèse porte sur la caractérisation de structures mécaniques, au travers de leurs paramètres structuraux, à partir d'observations perturbées par des bruits de mesure, supposés additifs blancs gaussiens et centrés. Pour cela, nous proposons d'utiliser des modèles à temps discret à parties linéaire et non-linéaire séparables. La première permet de retrouver les paramètres recherchés tandis que la seconde renseigne sur la non-linéarité présente. Dans le cadre d'une modélisation non-récursive par des séries de Volterra, nous présentons une approche à erreurs-dans-les-variables lorsque les variances des bruits ne sont pas connues ainsi qu'un algorithme adaptatif du type LMS nécessitant la connaissance de la variance du bruit d'entrée. Dans le cadre d'une modélisation par un modèle récursif polynomial, nous proposons deux méthodes à partir d'algorithmes évolutionnaires. La première inclut un protocole d'arrêt tenant compte de la variance du bruit de sortie. Dans la seconde, les fonctions fitness sont fondées sur des fonctions de corrélation dans lesquelles l'influence du bruit est supprimée ou compensée. / This PhD deals with the caracterisation of mechanical structures, by its structural parameters, when only noisy observations disturbed by additive measurement noises, assumed to be zero-mean white and Gaussian, are available. For this purpose, we suggest using discrete-time models with distinct linear and nonlinear parts. The first one allows the structural parameters to be retrieved whereas the second one gives information on the nonlinearity. When dealing with non-recursive Volterra series, we propose an errors-in-variables (EIV) method to jointly estimate the noise variances and the Volterra kernels. We also suggest a modified unbiased LMS algorithm to estimate the model parameters provided that the input-noise variance is known. When dealing with recursive polynomial model, we propose two methods using evolutionary algorithms. The first includes a stop protocol that takes into account the output-noise variance. In the second one, the fitness functions are based on correlation criteria in which the noise influence is removed or compensated.