Improved score tests in symmetric linear regression models

The class of symmetric linear regression models has the normal linear regression model as a special case and includes several models that assume that the errors follow a symmetric distribution with longer-than-normal tails. An important member of this class is the t linear regression model, which is commonly used as an alternative to the usual normal regression model when the data contain extreme or outlying observations. In this article, we develop second-order asymptotic theory for score tests in this class of models. We obtain Bartlett-corrected score statistics for testing hypotheses on the regression and the dispersion parameters. The corrected statistics have chi-squared distributions with errors of order O(n(-3/2)), n being the sample size. The corrections represent an improvement over the corresponding original Rao`s score statistics, which are chi-squared distributed up to errors of order O(n(-1)). Simulation results show that the corrected score tests perform much better than their uncorrected counterparts in samples of small or moderate size.

Methods of Estimation in Multiple Linear Regression: Application to Clinical Data

Nesse artigo, tem-se o interesse em avaliar diferentes estratégias de estimação de parâmetros para um modelo de regressão linear múltipla. Para a estimação dos parâmetros do modelo foram utilizados dados de um ensaio clínico em que o interesse foi verificar se o ensaio mecânico da propriedade de força máxima (EM-FM) está associada com a massa femoral, com o diâmetro femoral e com o grupo experimental de ratas ovariectomizadas da raça Rattus norvegicus albinus, variedade Wistar. Para a estimação dos parâmetros do modelo serão comparadas três metodologias: a metodologia clássica, baseada no método dos mínimos quadrados; a metodologia Bayesiana, baseada no teorema de Bayes; e o método Bootstrap, baseado em processos de reamostragem.

Bias correction in a multivariate normal regression model with general parameterization

This paper derives the second-order biases Of maximum likelihood estimates from a multivariate normal model where the mean vector and the covariance matrix have parameters in common. We show that the second order bias can always be obtained by means of ordinary weighted least-squares regressions. We conduct simulation studies which indicate that the bias correction scheme yields nearly unbiased estimators. (C) 2009 Elsevier B.V. All rights reserved.

The log-exponentiated Weibull regression model for interval-censored data

In interval-censored survival data, the event of interest is not observed exactly but is only known to occur within some time interval. Such data appear very frequently. In this paper, we are concerned only with parametric forms, and so a location-scale regression model based on the exponentiated Weibull distribution is proposed for modeling interval-censored data. We show that the proposed log-exponentiated Weibull regression model for interval-censored data represents a parametric family of models that include other regression models that are broadly used in lifetime data analysis. Assuming the use of interval-censored data, we employ a frequentist analysis, a jackknife estimator, a parametric bootstrap and a Bayesian analysis for the parameters of the proposed model. We derive the appropriate matrices for assessing local influences on the parameter estimates under different perturbation schemes and present some ways to assess global influences. Furthermore, for different parameter settings, sample sizes and censoring percentages, various simulations are performed; in addition, the empirical distribution of some modified residuals are displayed and compared with the standard normal distribution. These studies suggest that the residual analysis usually performed in normal linear regression models can be straightforwardly extended to a modified deviance residual in log-exponentiated Weibull regression models for interval-censored data. (C) 2009 Elsevier B.V. All rights reserved.

Deviance residuals in generalised log-gamma regression models with censored observations

In this article, we compare three residuals based on the deviance component in generalised log-gamma regression models with censored observations. For different parameter settings, sample sizes and censoring percentages, various simulation studies are performed and the empirical distribution of each residual is displayed and compared with the standard normal distribution. For all cases studied, the empirical distributions of the proposed residuals are in general symmetric around zero, but only a martingale-type residual presented negligible kurtosis for the majority of the cases studied. These studies suggest that the residual analysis usually performed in normal linear regression models can be straightforwardly extended for the martingale-type residual in generalised log-gamma regression models with censored data. A lifetime data set is analysed under log-gamma regression models and a model checking based on the martingale-type residual is performed.

Influence diagnostics in Birnbaum-Saunders nonlinear regression models

We consider the issue of assessing influence of observations in the class of Birnbaum-Saunders nonlinear regression models, which is useful in lifetime data analysis. Our results generalize those in Galea et al. [8] which are confined to Birnbaum-Saunders linear regression models. Some influence methods, such as the local influence, total local influence of an individual and generalized leverage are discussed. Additionally, the normal curvatures for studying local influence are derived under some perturbation schemes. We also give an application to a real fatigue data set.

Bias-corrected estimators for dispersion models with dispersion covariates

In this paper we discuss bias-corrected estimators for the regression and the dispersion parameters in an extended class of dispersion models (Jorgensen, 1997b). This class extends the regular dispersion models by letting the dispersion parameter vary throughout the observations, and contains the dispersion models as particular case. General formulae for the O(n(-1)) bias are obtained explicitly in dispersion models with dispersion covariates, which generalize previous results obtained by Botter and Cordeiro (1998), Cordeiro and McCullagh (1991), Cordeiro and Vasconcellos (1999), and Paula (1992). The practical use of the formulae is that we can derive closed-form expressions for the O(n(-1)) biases of the maximum likelihood estimators of the regression and dispersion parameters when the information matrix has a closed-form. Various expressions for the O(n(-1)) biases are given for special models. The formulae have advantages for numerical purposes because they require only a supplementary weighted linear regression. We also compare these bias-corrected estimators with two different estimators which are also bias-free to order O(n(-1)) that are based on bootstrap methods. These estimators are compared by simulation. (C) 2011 Elsevier B.V. All rights reserved.

Iron deficiency anaemia and blood lead concentrations in Brazilian children

This study investigated the relationship between iron deficiency/iron deficiency anaemia, assessed by several parameters, and blood lead concentration in children. This cross-sectional study involved 384 Brazilian children, aged 2-11 years, who lived near a lead-manipulating industry. Complete blood counts were obtained by an automated cell counter. Serum iron, total iron binding capacity (TIBC) and ferritin were determined respectively, by colorimetric, turbidimetric methods and chemiluminescence. Blood lead was measured by atomic absorption spectrophotometry. The impact of several parameters for assessment of iron status (haemoglobin, serum iron, TIBC, transferrin saturation, ferritin, red cell indices and red cell distribution width) and variables (gender, age, mother`s education, income, body mass index, iron intake, and distance from home to lead-manipulating industry) on blood lead concentration was determined by multiple linear regression. There were significant negative associations between blood lead and the distance from home to the lead-manipulating industry (P < 0.001), Hb (P = 0.019), and ferritin (P=0.023) (R(2)=0.14). Based on these results, further epidemiological studies are necessary to investigate the impact of interventions like iron supplementation or fortification, as an attempt to decrease blood lead in children. (C) 2011 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rights reserved.

Nutritional intervention programme among a Japanese-Brazilian community: procedures and results according to gender

Objective: To describe the results of a nutritional intervention programme among Japanese-Brazilians according to gender. Design: A non-controlled experimental study. Setting: The research included three points of clinical, nutritional and physical activity evaluation: at baseline (in 2005), after the first year and at the end of the second year (in 2007). The paired Student t test and multiple linear regression analysis were used to evaluate changes in the subjects` profile (clinical, nutritional and physical activity variables). Subjects: Japanese-Brazilians (n 575) of both genders, aged over 30 years. Results: We verified statistically significant reductions in body weight (0.9 kg), waist circumference (2.9 cm), blood pressure, fasting blood glucose (>3 mg/dl) and total cholesterol (>20 mg/dl) and its fractions, in both genders. We also found reductions in intake of energy (among men), protein (among women) and fat (both genders) and increases in intake of total fibre (among women) and carbohydrate (among men). Conclusions: The intervention programme indicated meaningful benefits for the intervention subjects, with changes in their habits that led to a `healthier` lifestyle positively impacting their nutritional and metabolic profile.

Dietary Quality and Associated Factors among Factory Workers in the Metropolitan Region of Sao Paulo, Brazil

The workplace is a manageable community-based setting for ensuring proper nutrition. This study aimed to evaluate dietary quality and associated factors among adult workers at a cosmetics factory in the metropolitan area of Sao Paulo, Brazil. This factory was actively participating in the Brazilian Workers` Meal Program, which was created to ensure workers` nutritional health. In this cross-sectional study, data on 202 adult workers were assessed using questionnaires (sociodemographic, anthropometric, and lifestyle characteristics) administered during August and September 2006. Dietary intake, measured by 24-hour dietary recall, was used to calculate the modified Healthy Eating Index (HEI). A repeated administration of the 24-hour dietary recall was applied in a random subsample to calculate the modified HEI adjusted for the within-person variation in intake. Mean adjusted modified HEI scores were analyzed using multiple linear regression adjusted for energy. The mean adjusted modified HEI score was 72.3 +/- 8.0. The lowest adjusted modified HEI components scores were ""milk and dairy products"" (4.4 +/- 3.2) and ""sodium"" (3.7 +/- 3.1). Two percent of workers had ""poor diet"" (adjusted modified HEI score <51 points) and the majority (87%) had ""diet that needs modification"" (adjusted modified HEI score between 51 and 80), despite their participation in the meal program. Adjusted modified HEI scores were considerably higher for men (74.7 +/- 7.0) than for women (66.9 +/- 8.2) and for normal body mass index (calculated as kg/m(2)) (73.3 +/- 7.8) than for overweight/obese (70.9 +/- 8.1). Based on these results, the vast majority of workers were found to have diets that needed improvement. Individuals with higher-quality diets were more likely to have lower body mass index and to be male. J Am Diet Assoc. 2010;110:786-790.

Dietary Quality Index and Associated Factors among Adolescents of the State of Sao Paulo, Brazil

Objective To assess dietary quality and associated factors in adolescents. Study design We conducted a population-based cross-sectional study in a sample of 1584 adolescents living in areas of the state of Sao Paulo, Brazil. Dietary intake was measured with the 24-hour recall method, and dietary quality was assessed by means of the Health Eating Index (HEI), adapted to fit to the local requirements. Linear regression analyses were performed to assess the association between the HEI and demographic, socioeconomic, and lifestyle variables. Results A total of 97.1% of the adolescents studied had an inadequate diet or a diet that needed improvement. The mean overall HEI score was 59.7. Lower mean HEI scores were found for fruits, dairy products, and vegetables. Male adolescents who were physically active and lived in a house or apartment had higher HEI scores. The multiple regression analyses showed that the quality of the diet improved as age decreased. Adolescents who lived in houses or apartments had higher HEI scores than adolescents living in shacks or slums, regardless of age and energy intake. Conclusions Dietary quality is associated with income and age. A better understanding of the factors associated can provide input to the formulation of policies and development of nutritional actions. (J Pediatr 2010; 156:456-60).

Dietary predictors of serum total carotene in low-income women living in Sao Paulo, south-east Brazil

Objective Dietary intake and nutritional status of antioxidant vitamins have been reported to protect against some cancers The objective of the present study was to assess the correlations between serum levels of carotenoids (including beta-, alpha- and gamma-carotene), lycopene, retinol, alpha- and gamma-tocopherols, and dietary intakes estimated by an FFQ, among low-income women in the Brazilian Investigation into Nutrition and Cervical Cancer Prevention (BRINCA) study. Design Cross-sectional study of data for 918 women aged 21-65 years participating in the BRINCA study in Sao Paulo city. Multiple linear regression models were used with serum nutrient levels as the dependent variable and dietary intake levels as the independent variable, adjusted for confounding factors. Results In energy-adjusted analyses, the intakes of dark green and deep yellow vegetables and fruits (partial R(2) = 4.8%), total fruits and juices (partial R(2) = 1.8%), vegetables and fruits (partial R(2) = 1.8%), carrots (partial R(2) = 1.4%) and citrus fruits and juices only (partial R(2) = 0.8%) were positively correlated only with serum total carotene levels, after adjusting for serum total cholesterol concentration, age, hospital attended, smoking status. BMI and presence of cervical lesions Multiple-adjusted serum levels of carotenoids were positively correlated with intake quartiles of dark green and deep yellow vegetables and fruits and total fruits and juices independent of smoking status. Conclusions The intake of specific fruits and vegetables was an independent predictor of serum total carotene levels in low-income women living in Sao Paulo

The influence of maternal factors on the concentration of vitamin A in mature breast milk

Background & aims: This study evaluated the relationship between vitamin A concentration in maternal milk and the characteristics of the donors of a Brazilian human milk bank. Material and methods: A total of 136 donors were selected in 2003-2004 for micronutrient determinations in breast milk and blood, anthropometric measurements and investigation of obstetric, socioeconomic-demographic factors, and life style. Maternal serum/milk samples were obtained for vitamin A, iron, copper, and zinc determinations. Vitamin A concentrations in breast milk and blood were assessed by high-performance liquid chromatography. Copper, zinc and iron concentrations in breast milk, and copper and zinc concentrations in blood were detected by atomic emission spectrophotometry. Serum ceruloplasmin and serum iron were determined, respectively, by nephelometry and colorimetry. A linear regression model assessed the associations between milk concentrations of vitamin A and maternal factors. Results: Vitamin A in milk presented positive associations with iron in milk (p < 0.001), serum retinol (p = 0.03), maternal work (p = 0.02), maternal age (p = 0.02). and oral contraceptive use (p = 0.01), and a negative association with % body fat (p = 0.01) (R(2) = 0.47). Conclusion: These results suggest that some nutritional, obstetric, and socioeconomic-demographic factors may have an effect on mature breast milk concentrations of vitamin A in apparently healthy Brazilian mothers. (C) 2009 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.

Dietary calcium intake and overweight: An epidemiologic view

Objective: We evaluated the relation between overweight and calcium intake in adults living in the municipality of Sao Paulo, Brazil. Methods: This was a cross-sectional population-based Study on a sample of 1459 adults that was obtained by multistage cluster sampling. Dietary intake was measured by the 24-h recall method. Poisson`s and linear regression analyses were performed to evaluate the relation between overweight and quartiles of calcium intake adjusted for energy. Results: The prevalence of overweight was 43.1% and the average adjusted calcium intake was 448.6 mg. In the linear regression. analyses, the regression coefficient for adjusted calcium was significant and negative (P = 0.019, beta(1) = -0.0001). Although evaluated by quartiles, the prevalence ratio for overweight in the first quartile of calcium intake was 1.24 (95% confidence interval 1.00-1.54) and that in the second quartile was 1.24 (95% confidence interval 1.03-1.49). Conclusion: In the present study, calcium intake showed a significant negative association with body mass index. (C) 2008 Elsevier Inc. All rights reserved.

High intake of fruits and vegetables predicts weight loss in Brazilian overweight adults

To determine whether changes in dietary intakes predict weight loss, we studied 80 overweight adults who attended a nutritional counseling program during 6 months of follow-up at a primary health care center in Brazil. Habitual diet was assessed using a validated food frequency questionnaire at baseline and after 6 months. The mean age (+/-SD) of the participants was 46.5 +/- 9.5 years, and their mean body mass index was 29 +/- 3 kg/m(2) at baseline. After 6 months, the differences in body weight and fruit/vegetable intake were -1.4 +/- 3 kg and 109 +/- 320 g daily, respectively. Using multiple linear regression models adjusted for age, sex, changes in walking time, and total energy intake, the increased intake of dietary fiber from fruits/vegetables was associated with a greater weight loss (beta 1 [95% confidence interval (CI)] = -0.180 [-0.269, -0.0911) after 6 months of follow-up. Similar results were observed for increased intake of vegetables (beta 1 [95% CI] = -0.00497 [-0.008, -0.0021) and fruits (beta 1 [95% CI] = -0.00290 [-0.005, -0.001]) as predictors of weight loss. The increase of 100 g/d of vegetables and fruits represented a body weight loss of 500 and 300 g after 6 months, respectively (P <.05). Our findings support the relevance of increased intakes of fruits and vegetables that may help avoid weight gain in overweight adults. (C) 2008 Elsevier Inc. All rights reserved.