Esta dissertação flexibiliza a suposição de normalidade, dispondo de distribuições assimétricas em modelos de crescimento. Propõe uma abordagem bayesiana para ajuste de modelos não lineares quando a suposição de normalidade para os erros não é razoável e/ou apresentam heteroscedasticidade. Assim, adota-se as distribuições skew-normal e skew-t para as situações em que é necessário modelar dados com caudas mais pesadas ou mais leves que a normal e assimétricos; sendo que é considerado também a presença de heteroscedasticidade. Diferentes funções são utilizadas na estrutura multiplicativa para modelar a variância. Com esse objetivo, métodos de inferência na abordagem bayesiana são desenvolvidos para estimar os parâmetros dos modelos de regressão não linear com os erros seguindo as distribuições citadas anteriormente. A metodologia visa aplicação à curvas de crescimento para dados de árvores / This paper relaxes the assumption of normality, featuring asymmetric distributions in growth models. Proposes a Bayesian approach to fit nonlinear models when the assumption of normality for the errors is not reasonable and/or exhibit heteroscedasticity. Thus, we adopt the skew-normal and skew-t distributions for situations where it is necessary to model data with tails heavier or lighter than normal and asymmetric, which is considered also the presence of heteroscedasticity. Different functions are used to model the multiplicative structure of variance. With this objective, methods of inference in the Bayesian approach are developed to estimate the parameters of nonlinear regression models with errors following the distributions listed above. The methodology is intended to apply to the growth curves for trees data sets

Accompanying the advances in computer technology is an increase collection of high dimensional data in many scientific and social studies. Sufficient dimension reduction (SDR) is a statistical method that enable us to reduce the dimension ofpredictors without loss of regression information. In this dissertation, we introduce principal asymmetric least squares (PALS) as a unified framework for linear and nonlinear sufficient dimension reduction. Classical methods such as sliced inverse regression (Li, 1991) and principal support vector machines (Li, Artemiou and Li, 2011) often do not perform well in the presence of heteroscedastic error, while our proposal addresses this limitation by synthesizing different expectile levels. Through extensive numerical studies, we demonstrate the superior performance of PALS in terms of both computation time and estimation accuracy. For the asymptotic analysis of PALS for linear sufficient dimension reduction, we develop new tools to compute the derivative of an expectation of a non-Lipschitz function. PALS is not designed to handle symmetric link function between the response and the predictors. As a remedy, we develop expectile-assisted inverse regression estimation (EA-IRE) as a unified framework for moment-based inverse regression. We propose to first estimate the expectiles through kernel expectile regression, and then carry out dimension reduction based on random projections of the regression expectiles. Several popular inverse regression methods in the literature including slice inverse regression, slice average variance estimation, and directional regression are extended under this general framework. The proposed expectile-assisted methods outperform existing moment-based dimension reduction methods in both numerical studies and an analysis of the Big Mac data. / Statistics

No description available.

No description available.

In simple linear regression, it is assumed that two variables are linearly related, with unknown intercept and slope parameters. In particular, a regressor variable is assumed to be precisely measurable, and a response is assumed to be a random variable whose mean depends on the regressor via a linear function. For the simple linear regression problem, interest typically centers on estimation of the unknown model parameters, and perhaps application of the resulting estimated linear relationship to make predictions about future response values corresponding to given regressor values. The linear statistical calibration problem (or, more precisely, the absolute linear calibration problem), bears a resemblance to simple linear regression. It is still assumed that the two variables are linearly related, with unknown intercept and slope parameters. However, in calibration, interest centers on estimating an unknown value of the regressor, corresponding to an observed value of the response variable. We consider Bayesian methods of analysis for the linear statistical calibration problem, based on noninformative priors. Posterior analyses are assessed and compared with classical inference procedures. It is shown that noninformative prior Bayesian analysis is a strong competitor, yielding posterior inferences that can, in many cases, be correctly interpreted in a frequentist context. We also consider extensions of the linear statistical calibration problem to polynomial models and multivariate regression models. For these models, noninformative priors are developed, and posterior inferences are derived. The results are illustrated with analyses of published data sets. In addition, a certain type of heteroscedasticity is considered, which relaxes the traditional assumptions made in the analysis of a statistical calibration problem. It is shown that the resulting analysis can yield more reliable results than an analysis of the homoscedastic model. / Ph. D.

We have seen an increased interest and extensive research in the monitoring of a process over time whose characteristics are represented mathematically in functional forms such as profiles. Most of the current techniques require all of the data for each profile to determine the state of the process. Thus, quality engineers from industrial processes such as agricultural, aquacultural, and chemical cannot make process corrections to the current profile that are essential for correcting their processes at an early stage. In addition, the focus of most of the current techniques is on the statistical significance of the parameters or features of the model instead of the practical significance, which often relates to the actual quality characteristic. The goal of this research is to provide alternatives to address these two main concerns. First, we study the use of a Shewhart type control chart to monitor within profiles, where the central line is the predictive mean profile and the control limits are formed based on the prediction band. Second, we study a statistic based on a non-linear mixed model recognizing that the model leads to correlations among the estimated parameters. / Doctor of Philosophy / Checking the stability over time of the quality of a process which is best expressed by a relationship between a quality characteristic and other variables involved in the process has received increasing attention. The goal of this research is to provide alternative methods to determine the state of such a process. Both methods presented here are compared to the current methodologies. The first method will allow us to monitor a process while the data is still being collected. The second one is based on the quality characteristic of the process and takes full advantage of the model structure. Both methods seem to be more robust than the current most well-known method.

Darbe siekiama aprašyti periodinį ilgos atminties finansinių laiko eilučių elgesį. Remiantis anksčiau sukurtais modeliais, siūlomas h-faktorių Gegenbauer-LARCH modelis, kuris į LARCH tipo proceso sąlyginės dispersijos lygtį įtraukia apibendrintą ilgos atminties filtrą, paremtą Gegenbauer polinomais. Darbe pateikiama anksčiau sukurtų modelių, skirtų finansinių aktyvų grąžų kintamumo modeliavimui, apžvalga. Remiantis ankstesnėmis idėjomis ir darbais, sukonstruojamas naujas Gegenbauer-LARCH modelis, kuriam tikrinama kovariacijos stacionarumo sąlyga. Pateikiamos modeliuotos h-faktorių Gegenbauer-LARCH proceso trajektorijos. Sukurtas modelis taikomas realiems Euro-Dolerio valiutų kurso duomenims. Identifikuotas modelio parametrai vertinami LUDE algoritmu, kuris maksimizuoja didžiausio tikėtinumo funkciją. Atliekama modelio adekvatumo analizė. Darbo pabaigoje pateikiamos išvados ir rekomendacijos. / On the ground of previous works and ideas a new class of models which describe long memory periodic behaviour in a time varying volatility of financial returns is introduced. Generalised periodic long-memory filters, based on Gegenbauer polynomials, are included into volatility equation of LARCH model and capture long memory periodic behaviour of the data. Thus, a new type of model called h-factor Gegenbauer-LARCH is presented. Moreover, a covariance stationarity condition is checked for one factor Gegenbauer-LARCH model. Also, generated processes are demonstrated. Furthermore, h-factor Gegenbauer-LARCH model is applied to Euro-Dollar hourly exchange rate returns. Identified model is estimated by means of LUDE algorithm which maximizes maximum likelihood function. The adequasy of the model is checked by reviewing residuals behaviour. Concerning empirical results the following conclusion is drawn: • Although model captures specific characteristics of the data such as slowly decaying periodic behaviour of autocorrelation function and pronounced peaks in periodogram but residuals analysis shows that model should be improved. Bordignon, Caporin, Lisi suggest that all possible frequencies were included to the model because higher frequencies might not be obvious from autocorrelation function or periodogram. However, we face computer capability problem. As a matter of fact, we cannot estimate a more complex model. Inclusion of autoregresive coefficients into the model did not... [to full text]

Classic parametric statistical tests, like the analysis of variance (ANOVA), are powerful tools used for comparing population means. These tests produce accurate results provided the data satisfies underlying assumptions such as homoscedasticity and balancedness, otherwise biased results are obtained. However, these assumptions are rarely satisfied in real-life. Alternative procedures must be explored. This thesis aims at investigating the impact of heteroscedasticity and unbalancedness on effect sizes in two-way fixed-effects ANOVA models. A real-life dataset, from which three different samples were simulated was used to investigate the changes in effect sizes under the influence of unequal variances and unbalancedness. The parametric bootstrap approach was proposed in case of unequal variances and non-normality. The results obtained indicated that heteroscedasticity significantly inflates effect sizes while unbalancedness has non-significant impact on effect sizes in two-way ANOVA models. However, the impact worsens when the data is both unbalanced and heteroscedastic. / Statistics / M. Sc. (Statistics)

No description available.

Em estudos longitudinais, repetidas observações de uma mesma variável resposta são coletadas na mesma unidade experimental, em diferentes ocasiões. Como diferentes observações são realizadas na mesma unidade, espera-se que estas sejam correlacionadas, e que exista uma heterogeneidade de variâncias nas diferentes ocasiões. Dados longitudinais multivariados são obtidos quando um conjunto de diferentes variáveis respostas são mensuradas na mesma unidade experimental repetidas vezes ao longo do tempo; nesse caso, além da correlação entre observações realizadas na mesma unidade experimental, deve-se considerar também a correlação entre diferentes variáveis respostas. Uma forma de analisar dados longitudinais bivariados é empregar um modelo misto para cada uma das variáveis respostas e uni-los em um modelo misto bivariado especificando a distribuição conjunta para os efeitos aleatórios. As estimativas dos parâmetros desta distribuição comum podem ser usadas para avaliar a relação entre as diferentes respostas. Para exemplificar a utilização da técnica, foram utilizados dados de armazenamento de leite UAT. Os modelos lineares mistos bivariados foram ajustados por meio do software SAS e a análise gráfica foi realizada por meio do software R. Para seleção dos modelos empregou-se os Critérios de Informação de Akaike (AIC) e Bayesiano (BIC), e o teste da razão de verossimilhanças para comparação de modelos encaixados. A utilização do modelo linear misto bivariado permitiu modelar a heterogeneidade de variâncias entre ocasiões e a correlação entre diferentes medidas na mesma unidade experimental, bem como a correlação entre as variáveis respostas. / In longitudinal studies, repeated measurements of a response variable are taken in the same experimental unit over time. . Since different observations are measured on the same experimental unit, it is expected that there is correlation among the repeated measurements and heterogeneity of variances in different occasions. Multivariate Longitudinal Data are obtained when we measure a number of different response variables in the same experimental unit repeatedly over time; in this case, we should also observe a correlation between the different response variables. One way to analyze bivariate longitudinal data is to use a mixed model for each of the response variables, and unite them in bivariate mixed models specifying the joint distribution for random effects. Parameter estimates of this common distribution may be used to evaluate the relationship between different responses. As an example of the use of the technique, UHT milk storage data were used. Models were fitted using SAS software and the graphical analysis was done with software R. To model selection, Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) were used, and maximum likelihood ratio test was used to compare nested models. The use of bivariate mixed linear model allowed to model the heteroscedasticity of the occasions, the correlation between the different measurements in the same experimental unit and also the correlation between the different response variables.