Performance of balanced two-stage empirical predictors of realized cluster latent values from finite populations: A simulation study

Predictors of random effects are usually based on the popular mixed effects (ME) model developed under the assumption that the sample is obtained from a conceptual infinite population; such predictors are employed even when the actual population is finite. Two alternatives that incorporate the finite nature of the population are obtained from the superpopulation model proposed by Scott and Smith (1969. Estimation in multi-stage surveys. J. Amer. Statist. Assoc. 64, 830-840) or from the finite population mixed model recently proposed by Stanek and Singer (2004. Predicting random effects from finite population clustered samples with response error. J. Amer. Statist. Assoc. 99, 1119-1130). Predictors derived under the latter model with the additional assumptions that all variance components are known and that within-cluster variances are equal have smaller mean squared error (MSE) than the competitors based on either the ME or Scott and Smith`s models. As population variances are rarely known, we propose method of moment estimators to obtain empirical predictors and conduct a simulation study to evaluate their performance. The results suggest that the finite population mixed model empirical predictor is more stable than its competitors since, in terms of MSE, it is either the best or the second best and when second best, its performance lies within acceptable limits. When both cluster and unit intra-class correlation coefficients are very high (e.g., 0.95 or more), the performance of the empirical predictors derived under the three models is similar. (c) 2007 Elsevier B.V. All rights reserved.

Local influence in estimating equations

Local influence diagnostics based on estimating equations as the role of a gradient vector derived from any fit function are developed for repeated measures regression analysis. Our proposal generalizes tools used in other studies (Cook, 1986: Cadigan and Farrell, 2002), considering herein local influence diagnostics for a statistical model where estimation involves an estimating equation in which all observations are not necessarily independent of each other. Moreover, the measures of local influence are illustrated with some simulated data sets to assess influential observations. Applications using real data are presented. (C) 2010 Elsevier B.V. All rights reserved.

Cutpoint selection for discretizing a continuous covariate for generalized estimating equations

We consider consider the problem of dichotomizing a continuous covariate when performing a regression analysis based on a generalized estimation approach. The problem involves estimation of the cutpoint for the covariate and testing the hypothesis that the binary covariate constructed from the continuous covariate has a significant impact on the outcome. Due to the multiple testing used to find the optimal cutpoint, we need to make an adjustment to the usual significance test to preserve the type-I error rates. We illustrate the techniques on one data set of patients given unrelated hematopoietic stem cell transplantation. Here the question is whether the CD34 cell dose given to patient affects the outcome of the transplant and what is the smallest cell dose which is needed for good outcomes. (C) 2010 Elsevier BM. All rights reserved.

Small-Sample Corrections for Score Tests in Birnbaum-Saunders Regressions

In this article, we deal with the issue of performing accurate small-sample inference in the Birnbaum-Saunders regression model, which can be useful for modeling lifetime or reliability data. We derive a Bartlett-type correction for the score test and numerically compare the corrected test with the usual score test and some other competitors.

Missing data mechanisms and their implications on the analysis of categorical data

We review some issues related to the implications of different missing data mechanisms on statistical inference for contingency tables and consider simulation studies to compare the results obtained under such models to those where the units with missing data are disregarded. We confirm that although, in general, analyses under the correct missing at random and missing completely at random models are more efficient even for small sample sizes, there are exceptions where they may not improve the results obtained by ignoring the partially classified data. We show that under the missing not at random (MNAR) model, estimates on the boundary of the parameter space as well as lack of identifiability of the parameters of saturated models may be associated with undesirable asymptotic properties of maximum likelihood estimators and likelihood ratio tests; even in standard cases the bias of the estimators may be low only for very large samples. We also show that the probability of a boundary solution obtained under the correct MNAR model may be large even for large samples and that, consequently, we may not always conclude that a MNAR model is misspecified because the estimate is on the boundary of the parameter space.

Measurement error models with a general class of error distribution

In general, the normal distribution is assumed for the surrogate of the true covariates in the classical error model. This paper considers a class of distributions, which includes the normal one, for the variables subject to error. An estimation approach yielding consistent estimators is developed and simulation studies reported.

Local Influence Under Parameter Constraints

Calculations of local influence curvatures and leverage have been well developed when the parameters are unrestricted. In this article, we discuss the assessment of local influence and leverage under linear equality parameter constraints with extensions to inequality constraints. Using a penalized quadratic function we express the normal curvature of local influence for arbitrary perturbation schemes and the generalized leverage matrix in interpretable forms, which depend on restricted and unrestricted components. The results are quite general and can be applied in various statistical models. In particular, we derive the normal curvature under three useful perturbation schemes for generalized linear models. Four illustrative examples are analyzed by the methodology developed in the article.

Systematic risk estimation in symmetric models

The aim of this article is to discuss the estimation of the systematic risk in capital asset pricing models with heavy-tailed error distributions to explain the asset returns. Diagnostic methods for assessing departures from the model assumptions as well as the influence of observations on the parameter estimates are also presented. It may be shown that outlying observations are down weighted in the maximum likelihood equations of linear models with heavy-tailed error distributions, such as Student-t, power exponential, logistic II, so on. This robustness aspect may also be extended to influential observations. An application in which the systematic risk estimate of Microsoft is compared under normal and heavy-tailed errors is presented for illustration.

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.

Gender and schooling inequalities in risk and protective factors for chronic diseases among Brazilian adults, through telephone survey

OBJECTIVES: To assess risk and protective factors for chronic noncommunicable diseases (CNCD) and to identify social inequalities in their distribution among Brazilian adults. METHODS: The data used were collected in 2007 through VIGITEL, an ongoing population-based telephone survey. This surveillance system was implemented in all of the Brazilian State capitals, over 54,000 interviews were analyzed. Age-adjusted prevalence ratios for trends at different schooling levels were calculated using Poisson regression with linear models. RESULTS: These analyses have shown differences in the prevalence of risk and protective factors for CNCD by gender and schooling. Among men, the prevalence ratios of overweight, consumption of meat with visible fat, and dyslipidemia were higher among men with more schooling, while tobacco use, sedentary lifestyle, and high-blood pressure were lower. Among women, tobacco use, overweight, obesity, high-blood pressure and diabetes were lower among men with more schooling, and consumption of meat with visible fat and sedentary lifestyles were higher. As for protective factors, fruit and vegetables intake and physical activity were higher in both men and women with more schooling. CONCLUSION: Gender and schooling influence on risk and protective factors for CNCD, being the values less favorable for men. vigitel is a useful tool for monitoring these factors amongst the Brazilian population.

Prevalence and social distribution of risk factors for chronic noncommunicable diseases in Brazil

OBJETIVOS: Evaluar los factores de riesgo de enfermedades crónicas no transmisibles (ECNT) e identificar las desigualdades sociales relacionadas con su distribución en la población adulta brasileña.MÉTODOS: Se estudiaron los factores de riesgo de ECNT (entre ellos el consumo de tabaco, el sobrepeso y la obesidad, el bajo consumo de frutas y vegetales [BCFV], la insuficiente actividad física en el tiempo de ocio [IAFTO], el estilo de vida sedentario y el consumo excesivo de alcohol) en una muestra probabilística de 54369 adultos de 26 capitales estatales de Brasil y el Distrito Federal en 2006. Se utilizó el Sistema de Vigilancia de los Factores Protectores y de Riesgo para Enfermedades Crónicas No Transmisibles por Entrevistas Telefónicas (VIGITEL), un sistema de encuestas telefónicas asistido por computadora, y se calcularon las prevalencias ajustadas por la edad para las tendencias en cuanto al nivel educacional mediante la regresión de Poisson con modelos lineales. RESULTADOS: Los hombres informaron mayor consumo de tabaco, sobrepeso, BCFV, estilo de vida sedentario y consumo excesivo de alcohol que las mujeres, pero menos IAFTO. En los hombres, la educación se asoció con un mayor sobrepeso y un estilo de vida sedentario, pero con un menor consumo de tabaco, BCFV e IAFTO. En las mujeres, la educación se asoció con un menor consumo de tabaco, sobrepeso, obesidad, BCFV e IAFTO, pero aumentó el estilo de vida sedentario CONCLUSIONES: En Brasil, la prevalencia de factores de riesgo para ECNT (excepto IAFTO) es mayor en los hombres que en las mujeres. En ambos sexos, el nivel de educación influye en la prevalencia de los factores de riesgo para ECNT

Inequalities in mortality of men by oral and pharyngeal cancer in Barcelona, Spain and Sao Paulo, Brazil, 1995-2003

Background: Large inequalities of mortality by most cancers in general, by mouth and pharynx cancer in particular, have been associated to behaviour and geopolitical factors. The assessment of socioeconomic covariates of cancer mortality may be relevant to a full comprehension of distal determinants of the disease, and to appraise opportune interventions. The objective of this study was to compare socioeconomic inequalities in male mortality by oral and pharyngeal cancer in two major cities of Europe and South America. Methods: The official system of information on mortality provided data on deaths in each city; general censuses informed population data. Age-adjusted death rates by oral and pharyngeal cancer for men were independently assessed for neighbourhoods of Barcelona, Spain, and Sao Paulo, Brazil, from 1995 to 2003. Uniform methodological criteria instructed the comparative assessment of magnitude, trends and spatial distribution of mortality. General linear models assessed ecologic correlations between death rates and socioeconomic indices (unemployment, schooling levels and the human development index) at the inner-city area level. Results obtained for each city were subsequently compared. Results: Mortality of men by oral and pharyngeal cancer ranked higher in Barcelona (9.45 yearly deaths per 100,000 male inhabitants) than in Spain and Europe as a whole; rates were on decrease. Sao Paulo presented a poorer profile, with higher magnitude (11.86) and stationary trend. The appraisal of ecologic correlations indicated an unequal and inequitably distributed burden of disease in both cities, with poorer areas tending to present higher mortality. Barcelona had a larger gradient of mortality than Sao Paulo, indicating a higher inequality of cancer deaths across its neighbourhoods. Conclusion: The quantitative monitoring of inequalities in health may contribute to the formulation of redistributive policies aimed at the concurrent promotion of wellbeing and social justice. The assessment of groups experiencing a higher burden of disease can instruct health services to provide additional resources for expanding preventive actions and facilities aimed at early diagnosis, standardized treatments and rehabilitation.

POLLING SYSTEMS WITH PARAMETER REGENERATION, THE GENERAL CASE

We consider a polling model with multiple stations, each with Poisson arrivals and a queue of infinite capacity. The service regime is exhaustive and there is Jacksonian feedback of served customers. What is new here is that when the server comes to a station it chooses the service rate and the feedback parameters at random; these remain valid during the whole stay of the server at that station. We give criteria for recurrence, transience and existence of the sth moment of the return time to the empty state for this model. This paper generalizes the model, when only two stations accept arriving jobs, which was considered in [Ann. Appl. Probab. 17 (2007) 1447-1473]. Our results are stated in terms of Lyapunov exponents for random matrices. From the recurrence criteria it can be seen that the polling model with parameter regeneration can exhibit the unusual phenomenon of null recurrence over a thick region of parameter space.

Performance of different immobilized lipases in palm oil biodiesel synthesis

Performance of different immobilized lipases in palm oil biodiesel synthesis. Optimized conditions for palm oil and ethanol enzymatic biodiesel synthesis were determined with different immobilized lipases SiO(2)-PVA-immobilized lipase from Pseudomonas fluorescens and acrylic resin-immobilized lipase, Novozym (R) 435, from Candida antartica, in solvent-free medium. A full factorial design assessed the influence of temperature (42 - 58 degrees C) and ethanol: palm oil (6:1 - 18:1) molar ratio on the transesterification yield. Main effects were adjusted by multiple regression analysis to linear models and the maximum transesterification yield was obtained at 42 degrees C and 18:1 ethanol: palm oil molar ratio. Mathematical models featuring total yield for each immobilized lipase were suitable to describe the experimental results.

The influence of environmental variables on capybara (Hydrochoerus hydrochaeris: Rodentia, Hydrochoeridae) detectability in anthropogenic environments of southeastern Brazil

Capybaras were monitored weekly from 1998 to 2006 by counting individuals in three anthropogenic environments (mixed agricultural fields, forest and open areas) of southeastern Brazil in order to examine the possible influence of environmental variables (temperature, humidity, wind speed, precipitation and global radiation) on the detectability of this species. There was consistent seasonality in the number of capybaras in the study area, with a specific seasonal pattern in each area. Log-linear models were fitted to the sample counts of adult capybaras separately for each sampled area, with an allowance for monthly effects, time trends and the effects of environmental variables. Log-linear models containing effects for the months of the year and a quartic time trend were highly significant. The effects of environmental variables on sample counts were different in each type of environment. As environmental variables affect capybara detectability, they should be considered in future species survey/monitoring programs.