Signed likelihood ratio tests in the Birnbaum-Saunders regression model

The Birnbaum-Saunders regression model is becoming increasingly popular in lifetime analyses and reliability studies. In this model, the signed likelihood ratio statistic provides the basis for testing inference and construction of confidence limits for a single parameter of interest. We focus on the small sample case, where the standard normal distribution gives a poor approximation to the true distribution of the statistic. We derive three adjusted signed likelihood ratio statistics that lead to very accurate inference even for very small samples. Two empirical applications are presented. (C) 2010 Elsevier B.V. All rights reserved.

Inference for a skew extension of the Grubbs model

In this paper, we discuss inferential aspects for the Grubbs model when the unknown quantity x (latent response) follows a skew-normal distribution, extending early results given in Arellano-Valle et al. (J Multivar Anal 96:265-281, 2005b). Maximum likelihood parameter estimates are computed via the EM-algorithm. Wald and likelihood ratio type statistics are used for hypothesis testing and we explain the apparent failure of the Wald statistics in detecting skewness via the profile likelihood function. The results and methods developed in this paper are illustrated with a numerical example.

A modified signed likelihood ratio test in elliptical structural models

In this paper we deal with the issue of performing accurate testing inference on a scalar parameter of interest in structural errors-in-variables models. The error terms are allowed to follow a multivariate distribution in the class of the elliptical distributions, which has the multivariate normal distribution as special case. We derive a modified signed likelihood ratio statistic that follows a standard normal distribution with a high degree of accuracy. Our Monte Carlo results show that the modified test is much less size distorted than its unmodified counterpart. An application is presented.

Bayesian Analysis for Nonlinear Regression Model Under Skewed Errors, with Application in Growth Curves

We have considered a Bayesian approach for the nonlinear regression model by replacing the normal distribution on the error term by some skewed distributions, which account for both skewness and heavy tails or skewness alone. The type of data considered in this paper concerns repeated measurements taken in time on a set of individuals. Such multiple observations on the same individual generally produce serially correlated outcomes. Thus, additionally, our model does allow for a correlation between observations made from the same individual. We have illustrated the procedure using a data set to study the growth curves of a clinic measurement of a group of pregnant women from an obstetrics clinic in Santiago, Chile. Parameter estimation and prediction were carried out using appropriate posterior simulation schemes based in Markov Chain Monte Carlo methods. Besides the deviance information criterion (DIC) and the conditional predictive ordinate (CPO), we suggest the use of proper scoring rules based on the posterior predictive distribution for comparing models. For our data set, all these criteria chose the skew-t model as the best model for the errors. These DIC and CPO criteria are also validated, for the model proposed here, through a simulation study. As a conclusion of this study, the DIC criterion is not trustful for this kind of complex model.

On estimation and influence diagnostics for the Grubbs` model under heavy-tailed distributions

The Grubbs` measurement model is frequently used to compare several measuring devices. It is common to assume that the random terms have a normal distribution. However, such assumption makes the inference vulnerable to outlying observations, whereas scale mixtures of normal distributions have been an interesting alternative to produce robust estimates, keeping the elegancy and simplicity of the maximum likelihood theory. The aim of this paper is to develop an EM-type algorithm for the parameter estimation, and to use the local influence method to assess the robustness aspects of these parameter estimates under some usual perturbation schemes, In order to identify outliers and to criticize the model building we use the local influence procedure in a Study to compare the precision of several thermocouples. (C) 2008 Elsevier B.V. All rights reserved.

Influence Diagnostics for a Skew Extension of the Grubbs Measurement Error Model

Influence diagnostics methods are extended in this article to the Grubbs model when the unknown quantity x (latent variable) follows a skew-normal distribution. Diagnostic measures are derived from the case-deletion approach and the local influence approach under several perturbation schemes. The observed information matrix to the postulated model and Delta matrices to the corresponding perturbed models are derived. Results obtained for one real data set are reported, illustrating the usefulness of the proposed methodology.

Modifications and Alternatives to the Tests of Levene and Brown & Forsythe for Equality of Variances and Means

The usual tests to compare variances and means (e. g. Bartlett`s test and F-test) assume that the sample comes from a normal distribution. In addition, the test for equality of means requires the assumption of homogeneity of variances. In some situation those assumptions are not satisfied, hence we may face problems like excessive size and low power. In this paper, we describe two tests, namely the Levene`s test for equality of variances, which is robust under nonnormality; and the Brown and Forsythe`s test for equality of means. We also present some modifications of the Levene`s test and Brown and Forsythe`s test, proposed by different authors. We analyzed and applied one modified form of Brown and Forsythe`s test to a real data set. This test is a robust alternative under nonnormality, heteroscedasticity and also when the data set has influential observations. The equality of variance can be well tested by Levene`s test with centering at the sample median.

Hypothesis testing in an errors-in-variables model with heteroscedastic measurement errors

In many epidemiological studies it is common to resort to regression models relating incidence of a disease and its risk factors. The main goal of this paper is to consider inference on such models with error-prone observations and variances of the measurement errors changing across observations. We suppose that the observations follow a bivariate normal distribution and the measurement errors are normally distributed. Aggregate data allow the estimation of the error variances. Maximum likelihood estimates are computed numerically via the EM algorithm. Consistent estimation of the asymptotic variance of the maximum likelihood estimators is also discussed. Test statistics are proposed for testing hypotheses of interest. Further, we implement a simple graphical device that enables an assessment of the model`s goodness of fit. Results of simulations concerning the properties of the test statistics are reported. The approach is illustrated with data from the WHO MONICA Project on cardiovascular disease. Copyright (C) 2008 John Wiley & Sons, Ltd.

A new class of survival regression models with heavy-tailed errors: robustness and diagnostics

Birnbaum-Saunders models have largely been applied in material fatigue studies and reliability analyses to relate the total time until failure with some type of cumulative damage. In many problems related to the medical field, such as chronic cardiac diseases and different types of cancer, a cumulative damage caused by several risk factors might cause some degradation that leads to a fatigue process. In these cases, BS models can be suitable for describing the propagation lifetime. However, since the cumulative damage is assumed to be normally distributed in the BS distribution, the parameter estimates from this model can be sensitive to outlying observations. In order to attenuate this influence, we present in this paper BS models, in which a Student-t distribution is assumed to explain the cumulative damage. In particular, we show that the maximum likelihood estimates of the Student-t log-BS models attribute smaller weights to outlying observations, which produce robust parameter estimates. Also, some inferential results are presented. In addition, based on local influence and deviance component and martingale-type residuals, a diagnostics analysis is derived. Finally, a motivating example from the medical field is analyzed using log-BS regression models. Since the parameter estimates appear to be very sensitive to outlying and influential observations, the Student-t log-BS regression model should attenuate such influences. The model checking methodologies developed in this paper are used to compare the fitted models.

A robust Bayesian approach to null intercept measurement error model with application to dental data

Measurement error models often arise in epidemiological and clinical research. Usually, in this set up it is assumed that the latent variable has a normal distribution. However, the normality assumption may not be always correct. Skew-normal/independent distribution is a class of asymmetric thick-tailed distributions which includes the Skew-normal distribution as a special case. In this paper, we explore the use of skew-normal/independent distribution as a robust alternative to null intercept measurement error model under a Bayesian paradigm. We assume that the random errors and the unobserved value of the covariate (latent variable) follows jointly a skew-normal/independent distribution, providing an appealing robust alternative to the routine use of symmetric normal distribution in this type of model. Specific distributions examined include univariate and multivariate versions of the skew-normal distribution, the skew-t distributions, the skew-slash distributions and the skew contaminated normal distributions. The methods developed is illustrated using a real data set from a dental clinical trial. (C) 2008 Elsevier B.V. All rights reserved.

Processes and variations in language economisation

This article analyses the processes of reducing language in textchats produced by non-native speakers of English. We propose that forms are reduced because of their high frequency and because of the discourse context. A wide variety of processes are attested in the literature, and we find different forms of clippings in our data, including mixtures of different clippings, homophone respellings, phonetic respellings including informal oral forms, initialisms (but no acronyms), and mixtures of clipping together with homophone and phonetic respellings. Clippings were the most frequent process (especially back-clippings and initialisms), followed by homophone respellings. There were different ways of metalinguistically marking reduction, but capitalisation was by far the most frequent. There is much individual variation in the frequencies of the different processes, although most were within normal distribution. The fact that nonnative speakers seem to generally follow reduction patterns of native speakers suggests that reduction is a universal process.

Fatores associados ao controle metabólico em pacientes com diabetes melito tipo 2 não usuários de insulina

Introdução: a obtenção de um bom controle metabólico é essencial para a prevenção das complicações crônicas do Diabetes Melito (DM). O tratamento é complexo e depende da implementação efetiva das diferentes estratégias terapêuticas disponíveis. Para que isso seja possível, é necessário que o paciente entenda os princípios terapêuticos e consiga executá-los. A precária educação em diabetes é percebida como um dos obstáculos para o alcance das metas terapêuticas. Objetivo: analisar, os fatores associados ao controle metabólico, em pacientes com DM tipo 2 (DM2) não usuários de insulina. Métodos: foi realizado um estudo transversal em pacientes com DM2 não usuários de insulina, selecionados ao acaso entre aqueles que consultavam nos ambulatórios de Medicina Interna, Endocrinologia e Enfermagem do Hospital de Clínicas de Porto Alegre. Os pacientes foram submetidos à avaliação clínica, laboratorial e responderam um questionário que incluía o tipo de tratamento realizado para DM, outros medicamentos e co-morbidades, pesquisa de complicações em ano prévio e avaliação do conhecimento sobre DM. Os pacientes foram classificados em dois grupos, com bom ou mau controle glicêmico, de acordo com o valor da glico-hemoglobina de 1 ponto % acima do limite superior do método utilizado. As comparações entre variáveis contínuas, com distribuição normal, foram analisadas pelo teste t de Student para amostras não-pareadas e para as variáveis de distribuição assimétrica ou com variância heterogênea o teste U de Mann-Whitney. A comparação entre percentagem foi feita pelo teste de qui-quadrado ou exato de Fisher. Foi realizada uma análise logística múltipla para identificar os fatores mais relevantes associados ao controle metabólico (variável dependente). As variáveis independentes com um nível de significância de P < 0,1 na análise bivariada, foram incluídas no modelo. Resultados: foram avaliados 143 pacientes com DM2, idade de 59,3 ± 10,1 anos, duração conhecida do DM 7,5 ± 6,3 anos, índice de massa corporal (IMC) de 29,7 ± 5,2 kg/m².Destes, 94 pacientes (65,73%) apresentavam bom controle glicêmico. Os pacientes com mau controle glicêmico usavam mais anti-hiperglicemiantes orais como monoterapia (OR = 9,37; IC = 2,60-33,81; P=0,004) ou associados (OR = 31,08; IC = 7,42-130,15; P < 0,001). Da mesma maneira, não fizeram dieta em dias de festa (OR = 3,29; IC = 1,51-7,16; P = 0,012). A inclusão do conhecimento sobre diabetes não foi diferente entre os pacientes com bom ou mau controle glicêmico (OR = 1,08; IC = 0,97 - 1,21; P = 0,219). A análise multivariada demonstrou que a consulta com a enfermeira educadora (OR = 0,24; IC = 0,108-0,534; P = 0,003), com o endocrinologista (OR = 0,15 ; IC = 0,063-0,373; P = 0,001) e o uso de hipolipemiantes (OR = 0,10; IC = 0,016 - 0,72; P = 0,054) foram associados ao bom controle glicêmico, ajustados para a não realização de dieta em festas, uso de anti-hiperglicemiantes orais e conhecimento sobre diabetes. Conclusão: o controle metabólico em pacientes DM2 é influenciado pelas atividades de educação com enfermeira e endocrinologista. O tratamento do DM2 deve incluir atividades de educação de forma sistemática.

Testing unit root based on partially adaptive estimation

This paper proposes unit tests based on partially adaptive estimation. The proposed tests provide an intermediate class of inference procedures that are more efficient than the traditional OLS-based methods and simpler than unit root tests based on fully adptive estimation using nonparametric methods. The limiting distribution of the proposed test is a combination of standard normal and the traditional Dickey-Fuller (DF) distribution, including the traditional ADF test as a special case when using Gaussian density. Taking into a account the well documented characteristic of heavy-tail behavior in economic and financial data, we consider unit root tests coupled with a class of partially adaptive M-estimators based on the student-t distributions, wich includes te normal distribution as a limiting case. Monte Carlo Experiments indicate that, in the presence of heavy tail distributions or innovations that are contaminated by outliers, the proposed test is more powerful than the traditional ADF test. We apply the proposed test to several macroeconomic time series that have heavy-tailed distributions. The unit root hypothesis is rejected in U.S. real GNP, supporting the literature of transitory shocks in output. However, evidence against unit roots is not found in real exchange rate and nominal interest rate even haevy-tail is taken into a account.

Cópulas: uma alternativa para a estimação de modelos de risco multivariados

Dentre os principais desafios enfrentados no cálculo de medidas de risco de portfólios está em como agregar riscos. Esta agregação deve ser feita de tal sorte que possa de alguma forma identificar o efeito da diversificação do risco existente em uma operação ou em um portfólio. Desta forma, muito tem se feito para identificar a melhor forma para se chegar a esta definição, alguns modelos como o Valor em Risco (VaR) paramétrico assumem que a distribuição marginal de cada variável integrante do portfólio seguem a mesma distribuição , sendo esta uma distribuição normal, se preocupando apenas em modelar corretamente a volatilidade e a matriz de correlação. Modelos como o VaR histórico assume a distribuição real da variável e não se preocupam com o formato da distribuição resultante multivariada. Assim sendo, a teoria de Cópulas mostra-se um grande alternativa, à medida que esta teoria permite a criação de distribuições multivariadas sem a necessidade de se supor qualquer tipo de restrição às distribuições marginais e muito menos as multivariadas. Neste trabalho iremos abordar a utilização desta metodologia em confronto com as demais metodologias de cálculo de Risco, a saber: VaR multivariados paramétricos - VEC, Diagonal,BEKK, EWMA, CCC e DCC- e VaR histórico para um portfólio resultante de posições idênticas em quatro fatores de risco – Pre252, Cupo252, Índice Bovespa e Índice Dow Jones

Comparação de carteiras otimizadas segundo o critério média-variância formadas através de estimativas robustas de risco e retorno

Este trabalho examinou as características de carteiras compostas por ações e otimizadas segundo o critério de média-variância e formadas através de estimativas robustas de risco e retorno. A motivação para isto é a distribuição típica de ativos financeiros (que apresenta outliers e mais curtose que a distribuição normal). Para comparação entre as carteiras, foram consideradas suas propriedades: estabilidade, variabilidade e os índices de Sharpe obtidos pelas mesmas. O resultado geral mostra que estas carteiras obtidas através de estimativas robustas de risco e retorno apresentam melhoras em sua estabilidade e variabilidade, no entanto, esta melhora é insuficiente para diferenciar os índices de Sharpe alcançados pelas mesmas das carteiras obtidas através de método de máxima verossimilhança para estimativas de risco e retorno.