Influence diagnostics in nonlinear mixed-effects elliptical models

In this work we propose and analyze nonlinear elliptical models for longitudinal data, which represent an alternative to gaussian models in the cases of heavy tails, for instance. The elliptical distributions may help to control the influence of the observations in the parameter estimates by naturally attributing different weights for each case. We consider random effects to introduce the within-group correlation and work with the marginal model without requiring numerical integration. An iterative algorithm to obtain maximum likelihood estimates for the parameters is presented, as well as diagnostic results based on residual distances and local influence [Cook, D., 1986. Assessment of local influence. journal of the Royal Statistical Society - Series B 48 (2), 133-169; Cook D., 1987. Influence assessment. journal of Applied Statistics 14 (2),117-131; Escobar, L.A., Meeker, W.Q., 1992, Assessing influence in regression analysis with censored data, Biometrics 48, 507-528]. As numerical illustration, we apply the obtained results to a kinetics longitudinal data set presented in [Vonesh, E.F., Carter, R.L., 1992. Mixed-effects nonlinear regression for unbalanced repeated measures. Biometrics 48, 1-17], which was analyzed under the assumption of normality. (C) 2009 Elsevier B.V. All rights reserved.

Power series generalized nonlinear models

We introduce in this paper a new class of discrete generalized nonlinear models to extend the binomial, Poisson and negative binomial models to cope with count data. This class of models includes some important models such as log-nonlinear models, logit, probit and negative binomial nonlinear models, generalized Poisson and generalized negative binomial regression models, among other models, which enables the fitting of a wide range of models to count data. We derive an iterative process for fitting these models by maximum likelihood and discuss inference on the parameters. The usefulness of the new class of models is illustrated with an application to a real data set. (C) 2008 Elsevier B.V. All rights reserved.

Improved likelihood inference for the shape parameter in Weibull regression

We obtain adjustments to the profile likelihood function in Weibull regression models with and without censoring. Specifically, we consider two different modified profile likelihoods: (i) the one proposed by Cox and Reid [Cox, D.R. and Reid, N., 1987, Parameter orthogonality and approximate conditional inference. Journal of the Royal Statistical Society B, 49, 1-39.], and (ii) an approximation to the one proposed by Barndorff-Nielsen [Barndorff-Nielsen, O.E., 1983, On a formula for the distribution of the maximum likelihood estimator. Biometrika, 70, 343-365.], the approximation having been obtained using the results by Fraser and Reid [Fraser, D.A.S. and Reid, N., 1995, Ancillaries and third-order significance. Utilitas Mathematica, 47, 33-53.] and by Fraser et al. [Fraser, D.A.S., Reid, N. and Wu, J., 1999, A simple formula for tail probabilities for frequentist and Bayesian inference. Biometrika, 86, 655-661.]. We focus on point estimation and likelihood ratio tests on the shape parameter in the class of Weibull regression models. We derive some distributional properties of the different maximum likelihood estimators and likelihood ratio tests. The numerical evidence presented in the paper favors the approximation to Barndorff-Nielsen`s adjustment.

On beta regression residuals

We propose two new residuals for the class of beta regression models, and numerically evaluate their behaviour relative to the residuals proposed by Ferrari and Cribari-Neto. Monte Carlo simulation results and empirical applications using real and simulated data are provided. The results favour one of the residuals we propose.

Improved likelihood inference in beta regression

We consider the issue of performing accurate small-sample likelihood-based inference in beta regression models, which are useful for modelling continuous proportions that are affected by independent variables. We derive small-sample adjustments to the likelihood ratio statistic in this class of models. The adjusted statistics can be easily implemented from standard statistical software. We present Monte Carlo simulations showing that inference based on the adjusted statistics we propose is much more reliable than that based on the usual likelihood ratio statistic. A real data example is presented.

Framework for Skew-Probit Links in Binary Regression

We review several asymmetrical links for binary regression models and present a unified approach for two skew-probit links proposed in the literature. Moreover, under skew-probit link, conditions for the existence of the ML estimators and the posterior distribution under improper priors are established. The framework proposed here considers two sets of latent variables which are helpful to implement the Bayesian MCMC approach. A simulation study to criteria for models comparison is conducted and two applications are made. Using different Bayesian criteria we show that, for these data sets, the skew-probit links are better than alternative links proposed in the literature.

Diagnostic tools in beta regression with varying dispersion

We consider the issue of performing residual and local influence analyses in beta regression models with varying dispersion, which are useful for modelling random variables that assume values in the standard unit interval. In such models, both the mean and the dispersion depend upon independent variables. We derive the appropriate matrices for assessing local influence on the parameter estimates under different perturbation schemes. An application using real data is presented and discussed.

Regression analysis of mean quality-adjusted survival time based on pseudo-observations

Regression models for the mean quality-adjusted survival time are specified from hazard functions of transitions between two states and the mean quality-adjusted survival time may be a complex function of covariates. We discuss a regression model for the mean quality-adjusted survival (QAS) time based on pseudo-observations, which has the advantage of directly modeling the effect of covariates in the QAS time. Both Monte Carlo Simulations and a real data set are studied. Copyright (C) 2009 John Wiley & Sons, Ltd.

Influence diagnostics for linear models with first-order autoregressive elliptical errors

We introduce in this paper the class of linear models with first-order autoregressive elliptical errors. The score functions and the Fisher information matrices are derived for the parameters of interest and an iterative process is proposed for the parameter estimation. Some robustness aspects of the maximum likelihood estimates are discussed. The normal curvatures of local influence are also derived for some usual perturbation schemes whereas diagnostic graphics to assess the sensitivity of the maximum likelihood estimates are proposed. The methodology is applied to analyse the daily log excess return on the Microsoft whose empirical distributions appear to have AR(1) and heavy-tailed errors. (C) 2008 Elsevier B.V. All rights reserved.

Influence diagnostics in beta regression

We consider the issue of assessing influence of observations in the class of beta regression models, which is useful for modelling random variables that assume values in the standard unit interval and are affected by independent variables. We propose a Cook-like distance and also measures of local influence under different perturbation schemes. Applications using real data are presented. (c) 2008 Elsevier B.V.. All rights reserved.

Inequalities in the distribution of dental caries among 12-year-old Brazilian schoolchildren

We assessed the inequality in the distribution of dental caries and the association between indicators of socioeconomic status and caries experience in a representative sample of schoolchildren. This study followed a cross-sectional design, with a sample of 792 schoolchildren aged 12 years, representative of this age group in Santa Maria, RS, Brazil. Guardians answered questions on socioeconomic status and a dental examination provided information on the dental caries experience (DMF-T). Inequality in dental caries distribution was measured by the Gini coefficient and the Significant Caries Index (SiC). The assessment of association used Poisson regression models. Socioeconomic factors were associated with prevalence of dental caries for the whole sample and also for individuals with a high-caries level. Children from low-income households had the highest prevalence of dental caries. The Gini coefficient was 0.7 and the SiC Index 2.5. The percentage of caries prevalence was 39.3% (95% CI: 35.8%-42.8%) and the mean for DMF-T was 0.9 (± SD 1.5). Inequalities in the distribution of dental caries were observed and socioeconomic factors were found to be strong predictors of the prevalence of oral disease in children of this age group.

Modelos preditores da fitomassa aérea da Floresta Baixa de Restinga

Este estudo teve como objetivo desenvolver modelos preditores de fitomassa epigéa da vegetação arbórea da Floresta Baixa de Restinga. Foram selecionadas 102 árvores de 29 espécies ocorrentes na área de estudo e 102 indivíduos de jerivá (Syagrus romanzoffiana (Cham.) Glassman), distribuídos proporcionalmente entre as classes de diâmetro da vegetação arbórea. As árvores foram cortadas, ao nível do solo e foram medidos a altura total e o diâmetro à altura do peito (DAP) de cada árvore. As folhas foram separadas do lenho e a massa fresca da porção lenhosa e foliar medidas separadamente. Amostras de cada fração foram secas a 70 °C, até peso constante, para determinação da massa seca das árvores. Os modelos foram desenvolvidos através de análise de regressão linear, sendo a variável dependente a massa seca (MS) das árvores e as variáveis independentes a altura (h), o diâmetro a altura do peito (d) e as relações d² h e d² h multiplicada pela densidade da madeira (ρ d² h). Os modelos desenvolvidos indicam que o diâmetro explica grande parte da variabilidade da fitomassa das árvores da restinga e a altura é a variável explanatória da equação específica para o jerivá. Os modelos selecionados foram: ln MS (kg) = -1,352 + 2,009 ln d (R² = 0,96; s yx = 0,34) para a comunidade vegetal sem jerivá, ln MS (kg) = -2,052 + 0,801 ln d² h (R² = 0,94; s yx = 0,38) para a comunidade incluindo o jerivá, e ln MS (kg) = -0,884 + 2,40 ln h (R² = 0,92; s yx = 0,49) para o jerivá.

Alcohol dehydrogenase activities and ethanol tolerance in Anastrepha (Diptera, Tephritidae) fruit-fly species and their hybrids

The ADH (alcohol dehydrogenase) system is one of the earliest known models of molecular evolution, and is still the most studied in Drosophila. Herein, we studied this model in the genus Anastrepha (Diptera, Tephritidae). Due to the remarkable advantages it presents, it is possible to cross species with different Adh genotypes and with different phenotype traits related to ethanol tolerance. The two species studied here each have a different number of Adh gene copies, whereby crosses generate polymorphisms in gene number and in composition of the genetic background. We measured certain traits related to ethanol metabolism and tolerance. ADH specific enzyme activity presented gene by environment interactions, and the larval protein content showed an additive pattern of inheritance, whilst ADH enzyme activity per larva presented a complex behavior that may be explained by epistatic effects. Regression models suggest that there are heritable factors acting on ethanol tolerance, which may be related to enzymatic activity of the ADHs and to larval mass, although a pronounced environmental effect on ethanol tolerance was also observed. By using these data, we speculated on the mechanisms of ethanol tolerance and its inheritance as well as of associated traits.

Impact on human health of particulate matter emitted from burnings in the Brazilian Amazon region

OBJECTIVE: To analyze the impact on human health of exposure to particulate matter emitted from burnings in the Brazilian Amazon region. METHODS: This was an ecological study using an environmental exposure indicator presented as the percentage of annual hours (AH%) of PM2.5 above 80 μg/m3. The outcome variables were the rates of hospitalization due to respiratory disease among children, the elderly and the intermediate age group, and due to childbirth. Data were obtained from the National Space Research Institute and the Ministry of Health for all of the microregions of the Brazilian Amazon region, for the years 2004 and 2005. Multiple regression models for the outcome variables in relation to the predictive variable AH% of PM2.5 above 80 μg/m3 were analyzed. The Human Development Index (HDI) and mean number of complete blood counts per 100 inhabitants in the Brazilian Amazon region were the control variables in the regression analyses. RESULTS: The association of the exposure indicator (AH%) was higher for the elderly than for other age groups (β = 0.10). For each 1% increase in the exposure indicator there was an increase of 8% in child hospitalization, 10% in hospitalization of the elderly, and 5% for the intermediate age group, even after controlling for HDI and mean number of complete blood counts. No association was found between the AH% and hospitalization due to childbirth. CONCLUSIONS: The indicator of atmospheric pollution showed an association with occurrences of respiratory diseases in the Brazilian Amazon region, especially in the more vulnerable age groups. This indicator may be used to assess the effects of forest burning on human health.

Tendência temporal e desigualdades na mortalidade por diarreias em menores de 5 anos

OBJETIVO: Analisar a tendência da mortalidade por diarreia entre menores de 5 anos, no município de Osasco (SP), entre 1980 e 2000. MÉTODOS: Trata-se de estudo observacional com dois delineamentos. Um descritivo, que toma o indivíduo como unidade do estudo, e outro ecológico, analisando agregado populacional que incluiu análise de séries temporais. A fonte de dados foi o sistema de informação de mortalidade do Estado de São Paulo e censos de 1980, 1991 e 2000. Descreveu-se a variação sazonal e para a análise de tendência aplicaram-se modelos log lineares de regressão polinomiais, utilizando-se variáveis sociodemográficas da criança e da mãe. Foram analisadas a evolução de indicadores sociodemográficos do município de 1980 a 2000, as taxas médias de mortalidade por diarreia nos menores de 5 anos e seus diferenciais por distrito nos anos 90. RESULTADOS: Dos 1.360 óbitos, 94,3 e 75,3% atingiram, respectivamente, menores de 1 ano e de 6 meses. O declínio da mortalidade foi de 98,3%, com deslocamento da sazonalidade do verão para o outono. A mediana da idade elevou-se de 2 meses nos primeiros períodos para 3 meses no último. O resíduo de óbitos manteve-se entre filhos de mães de 20 a 29 anos e escolaridade < 8 anos. O risco relativo entre o distrito mais atingido e a taxa média do município diminuiu de 3,4 para 1,3 do primeiro para o segundo quinquênio dos anos 90. CONCLUSÃO: Nossos resultados apontam uma elevação da idade mais vulnerável e a provável mudança do agente mais frequentemente associado ao óbito por diarreia.