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.

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.

Optimization of the rheological properties of probiotic yoghurts supplemented with milk proteins

This study aimed to optimize the rheological properties of probiotic yoghurts supplemented with skimmed milk powder (SMP) whey protein concentrate (WPC) and sodium caseinate (Na-Cn) by using an experimental design type simplex-centroid for mixture modeling It Included seven batches/trials three were supplemented with each type of the dairy protein used three corresponding to the binary mixtures and one to the ternary one in order to increase protein concentration in 1 g 100 g(-1) of final product A control experiment was prepared without supplementing the milk base Processed milk bases were fermented at 42 C until pH 4 5 by using a starter culture blend that consisted of Streptococcus thermophilus Lactobacillus delbrueckii subsp bulgaricus and Bifidobacterium (Humans subsp lactis The kinetics of acidification was followed during the fermentation period as well the physico-chemical analyses enumeration of viable bacteria and theological characteristics of the yoghurts Models were adjusted to the results (kinetic responses counts of viable bacteria and theological parameters) through three regression models (linear quadratic and cubic special) applied to mixtures The results showed that the addition of milk proteins affected slightly acidification profile and counts of S thermophilus and B animal`s subsp lactis but it was significant for L delbrueckii subsp bulgaricus Partially-replacing SMP (45 g/100 g) with WPC or Na-Cn simultaneously enhanced the theological properties of probiotic yoghurts taking into account the kinetics of acidification and enumeration of viable bacteria (C) 2010 Elsevier Ltd All rights reserved

A Note on Proxy Variables and Instrument Variable Regression

This is a note about proxy variables and instruments for identification of structural parameters in regression models. We have experienced that in the econometric textbooks these two issues are treated separately, although in practice these two concepts are very often combined. Usually, proxy variables are inserted in instrument variable regressions with the motivation they are exogenous. Implicitly meaning they are exogenous in a reduced form model and not in a structural model. Actually if these variables are exogenous they should be redundant in the structural model, e.g. IQ as a proxy for ability. Valid proxies reduce unexplained variation and increases the efficiency of the estimator of the structural parameter of interest. This is especially important in situations when the instrument is weak. With a simple example we demonstrate what is required of a proxy and an instrument when they are combined. It turns out that when a researcher has a valid instrument the requirements on the proxy variable is weaker than if no such instrument exists

Speed prediction for triggering vehicle activated signs

Accurate speed prediction is a crucial step in the development of a dynamic vehcile activated sign (VAS). A previous study showed that the optimal trigger speed of such signs will need to be pre-determined according to the nature of the site and to the traffic conditions. The objective of this paper is to find an accurate predictive model based on historical traffic speed data to derive the optimal trigger speed for such signs. Adaptive neuro fuzzy (ANFIS), classification and regression tree (CART) and random forest (RF) were developed to predict one step ahead speed during all times of the day. The developed models were evaluated and compared to the results obtained from artificial neural network (ANN), multiple linear regression (MLR) and naïve prediction using traffic speed data collected at four sites located in Sweden. The data were aggregated into two periods, a short term period (5-min) and a long term period (1-hour). The results of this study showed that using RF is a promising method for predicting mean speed in the two proposed periods.. It is concluded that in terms of performance and computational complexity, a simplistic input features to the predicitive model gave a marked increase in the response time of the model whilse still delivering a low prediction error.

Aplicação de redes neurais na classificação de rentabilidade futura de empresas

Este trabalho tem por motivação evidenciar a eficiência de redes neurais na classificação de rentabilidade futura de empresas, e desta forma, prover suporte para o desenvolvimento de sistemas de apoio a tomada de decisão de investimentos. Para serem comparados com o modelo de redes neurais, foram escolhidos o modelo clássico de regressão linear múltipla, como referência mínima, e o de regressão logística ordenada, como marca comparativa de desempenho (benchmark). Neste texto, extraímos dados financeiros e contábeis das 1000 melhores empresas listadas, anualmente, entre 1996 e 2006, na publicação Melhores e Maiores – Exame (Editora Abril). Os três modelos foram construídos tendo como base as informações das empresas entre 1996 e 2005. Dadas as informações de 2005 para estimar a classificação das empresas em 2006, os resultados dos três modelos foram comparados com as classificações observadas em 2006, e o modelo de redes neurais gerou o melhor resultado.

Essays in Corporate Bankruptcy

Esta tese é composta por três ensaios sobre o mercado de crédito e as instituições que regem bancarrota corporativa. No capítulo um, trazemos evidências que questionam a ideia de que maiores níveis de proteção ao credor sempre promovem desenvolvimento do mercado de crédito. Desde a publicação dos artigos seminais de La Porta et al (1997,1998), a métrica de proteção ao credor que os autores propuseram -- o índice de proteção ao credor -- tem sido amplamente utilizada na literatura de Law and Finance como variável explicativa em modelos de regressão linear em forma reduzida para determinar a correlação entre proteção ao credor e desenvolvimento do mercado de crédito. Neste artigo, exploramos alguns problemas com essa abordagem. Do ponto de vista teórico, essa abordagem geralmente supõe uma relação monotônica entre proteção ao credor e expansão do crédito. Nós apresentamos um modelo teórico para um mercado de crédito com seleção adversa em que um nível intermediário de proteção ao credor é capaz de implementar equilíbrios first best. Este resultado está de acordo com diversos outros artigos teóricos, tanto em equilíbrio geral quanto em equilíbrio parcial. Do ponto de vista empírico, tiramos proveito das reformas realizadas por alguns países durante as décadas de 1990 e 2000 para implementar uma estratégia inspirada na literatura de treatment effects e estimar o efeito sobre o valor de mercado e sobre a dívida de: i) permitir automatic stay a firmas em recuperação; e ii) conceder aos credores o direito de afastar os administradores. Os resultados que obtivemos apontam para um impacto positivo de automatic stay sobre todas as variáveis que dependem do valor de mercado da firma. Não encontramos efeito sobre dívida, e não encontramos efeitos significativos do direito de afastar administradores sobre valor de mercado ou dívida. O capítulo dois avalia as consequências empíricas de uma reforma na lei de falências sobre um mercado de crédito pouco desenvolvido. No início de 2005, o Congresso Nacional brasileiro aprovou uma nova lei de falências, a lei 11.101/05. Usando dados de firmas brasileiras e não-brasileiras, nós estimamos, usando dois modelos diferentes, o efeito da reforma falimentar sobre variáveis contratuais e não-contratuais de dívida. Ambos os modelos produzem resultados similares. Encontramos um aumento no volume total de dívida e na dívida de longo prazo, e uma redução no custo de dívida. Não encontramos efeitos significativos sobre a estrutura de propriedade da dívida. No capítulo três, desenvolvemos um modelo estimável de equilíbrio em search direcionado aplicado ao mercado de crédito, modelo este que pode ser usado para realizar avaliações ex ante de mudanças institucionais que afetem o crédito (como reformas em leis de falência). A literatura em economia há muito reconhece uma relação causal entre instituições (como leis e regulações) e desenvolvimento dos mercados financeiros. Essa conclusão qualitativa é amplamente reconhecida, mas há pouca evidência de sua importância quantitativa. Com o nosso modelo, é possível estimar como contratos de dívida mudam em resposta a mudanças nos parâmetros que descrevem as instituições da economia. Também é possível estimar o impacto sobre investimentos realizados pelas firmas, bem como caracterizar a distribuição do tamanho, idade e produtividade das firmas antes e depois da mudança institucional. Como ilustração, realizamos um exercício empírico em que usamos dados de firmas brasileiras para simular o impacto de variações na taxa de recuperação de créditos sobre os valores médios e totais de dívida e capital das firmas. Encontramos dívida crescente e capital quase sempre também crescente na taxa de recuperação.

Improved combinations of parametric and nonparametric Regression estimators

This paper provides a systematic and unified treatment of the developments in the area of kernel estimation in econometrics and statistics. Both the estimation and hypothesis testing issues are discussed for the nonparametric and semiparametric regression models. A discussion on the choice of windowwidth is also presented.

Tree-structured smooth transition models

The goal of this paper is to introduce a class of tree-structured models that combines aspects of regression trees and smooth transition regression models. The model is called the Smooth Transition Regression Tree (STR-Tree). The main idea relies on specifying a multiple-regime parametric model through a tree-growing procedure with smooth transitions among different regimes. Decisions about splits are entirely based on a sequence of Lagrange Multiplier (LM) tests of hypotheses.

Eficiência e razão de hedge: uma análise dos mercados futuro brasileiros de boi, café, etanol, milho e soja

This research aims to investigate the Hedge Efficiency and Optimal Hedge Ratio for the future market of cattle, coffee, ethanol, corn and soybean. This paper uses the Optimal Hedge Ratio and Hedge Effectiveness through multivariate GARCH models with error correction, attempting to the possible phenomenon of Optimal Hedge Ratio differential during the crop and intercrop period. The Optimal Hedge Ratio must be bigger in the intercrop period due to the uncertainty related to a possible supply shock (LAZZARINI, 2010). Among the future contracts studied in this research, the coffee, ethanol and soybean contracts were not object of this phenomenon investigation, yet. Furthermore, the corn and ethanol contracts were not object of researches which deal with Dynamic Hedging Strategy. This paper distinguishes itself for including the GARCH model with error correction, which it was never considered when the possible Optimal Hedge Ratio differential during the crop and intercrop period were investigated. The commodities quotation were used as future price in the market future of BM&FBOVESPA and as spot market, the CEPEA index, in the period from May 2010 to June 2013 to cattle, coffee, ethanol and corn, and to August 2012 to soybean, with daily frequency. Similar results were achieved for all the commodities. There is a long term relationship among the spot market and future market, bicausality and the spot market and future market of cattle, coffee, ethanol and corn, and unicausality of the future price of soybean on spot price. The Optimal Hedge Ratio was estimated from three different strategies: linear regression by MQO, BEKK-GARCH diagonal model, and BEKK-GARCH diagonal with intercrop dummy. The MQO regression model, pointed out the Hedge inefficiency, taking into consideration that the Optimal Hedge presented was too low. The second model represents the strategy of dynamic hedge, which collected time variations in the Optimal Hedge. The last Hedge strategy did not detect Optimal Hedge Ratio differential between the crop and intercrop period, therefore, unlikely what they expected, the investor do not need increase his/her investment in the future market during the intercrop

Estado nutricional antropométrico, perfil lipídico e consumo alimentar como fatores de risco para doença cardiovascular em adolescentes

The dyslipidemia and excess weight in adolescents, when combined, suggest a progression of risk factors for cardiovascular disease (CVD). Besides these, the dietary habits and lifestyle have also been considered unsuitable impacting the development of chronic diseases. The study objectives were: (1) estimate the prevalence of lipid profile and correlate with body mass index (BMI), waist circumference (WC) and waist / height ratio (WHR) in adolescents, considering the maturation sexual, (2) know the sources of variance in the diet and the number of days needed to estimate the usual diet of adolescents and (3) describe the dietary patterns and lifestyle of adolescents, family history of CVD and age correlates them with the patterns of risk for CVD, adjusted for sexual maturation. A cross-sectional study was performed with 432 adolescents, aged 10-19 years from public schools of the Natal city, Brazil. The dyslipidemias were evaluated considering the lipid profile, the index of I Castelli (TC / HDL) and II (LDL / HDL) and non-HDL cholesterol. Anthropometric indicators were BMI, WC and WHR. The intake of energy, nutrients including fiber, fatty acids and cholesterol was estimated from two 24-hour recalls (24HR). The variables of lipid profile, anthropometric and clinical data were used in the models of Pearson correlation and linear regression, considering the sexual maturation. The variance ratio of the diet was calculated from the component-person variance, determined by analysis of variance (ANOVA). The definition of the number of days to estimate the usual intake of each nutrient was obtained by taking the hypothetical correlation (r) ≥ 0.9, between nutrient intake and the true observed. We used the principal component analysis as a method of extracting factors that 129 accounted for the dependent variables and known cardiovascular risk obtained from the lipid profile, the index for Castelli I and II, non-HDL cholesterol, BMI, and WC the WHR. Dietary patterns and lifestyle were obtained from the independent variables, based on nutrients consumed and physical activity weekly. In the study of principal component analysis (PCA) was investigated associations between the patterns of cardiovascular risk factors in dietary patterns and lifestyle, age and positive family history of CVD, through bivariate and multiple logistic regression adjusted for sexual maturation. The low HDL-C dyslipidemia was most prevalent (50.5%) for adolescents. Significant correlations were observed between hypercholesterolemia and positive family history of CVD (r = 0.19, p <0.01) and hypertriglyceridemia with BMI (r = 0.30, p <0.01), with the CC (r = 0.32, p <0.01) and WHR (r = 0.33, p <0.01). The linear model constructed with sexual maturation, age and BMI explained about 1 to 10.4% of the variation in the lipid profile. The sources of variance between individuals were greater for all nutrients in both sexes. The reasons for variances were  1 for all nutrients were higher in females. The results suggest that to assess the diet of adolescents with greater precision, 2 days would be enough to R24h consumption of energy, carbohydrates, fiber, saturated and monounsaturated fatty acids. In contrast, 3 days would be recommended for protein, lipid, polyunsaturated fatty acids and cholesterol. Two cardiovascular risk factors as have been extracted in the ACP, referring to the dependent variables: the standard lipid profile (HDL-C and non-HDL cholesterol) and "standard anthropometric index (BMI, WC, WHR) with a power explaining 75% of the variance of the original data. The factors are representative of two independent variables led to dietary patterns, "pattern 130 western diet" and "pattern protein diet", and one on the lifestyle, "pattern energy balance". Together, these patterns provide an explanation power of 67%. Made adjustment for sexual maturation in males remained significant variables: the associations between puberty and be pattern anthropometric indicator (OR = 3.32, CI 1.34 to 8.17%), and between family history of CVD and the pattern lipid profile (OR = 2.62, CI 1.20 to 5.72%). In females adolescents, associations were identified between age after the first stage of puberty with anthropometric pattern (OR = 3.59, CI 1.58 to 8.17%) and lipid profile (OR = 0.33, CI 0.15 to 0.75%). Conclusions: The low HDL-C was the most prevalent dyslipidemia independent of sex and nutritional status of adolescents. Hypercholesterolemia was influenced by family history of CVD and sexual maturation, in turn, hypertriglyceridemia was closely associated with anthropometric indicators. The variance between the diets was greater for all nutrients. This fact reflected in a variance ratio less than 1 and consequently in a lower number of days requerid to estimate the usual diet of adolescents considering gender. The two dietary patterns were extracted and the pattern considered unhealthy lifestyle as healthy. The associations were found between the patterns of CVD risk with age and family history of CVD in the studied adolescents

Estimativa da área foliar de Ageratum conyzoides usando dimensões lineares do limbo foliar

The aim of this research was to obtain a mathematical equation to estimate the leaf area of Ageratum conyzoides based on linear measures of its leaf blade. Correlation studies were done using real leaf area (Sf), leaf length (C) and the maximum leaf width (L), in about 200 leaf blades. The evaluated statistic models were: linear Y = a + bx; simple linear Y = bx; geometric Y = ax(b); and exponential Y = ab(x). The evaluated linear, exponential and geometric models can be used in the billygoat weed leaf area estimation. In the practical sense, the simple linear regression model is suggested using the C*L multiplication product and taking the linear coefficient equal to zero, because it showed weak-alteration on sum of squares error and satisfactory residual analysis. Thus, an estimate of A conyzoides leaf area can be obtained using the equation Sf = 0.6789*(C*L), with a determination coefficient of 0.8630.

Estimativa da área foliar de Sida cordifolia e Sida rhombifolia usando dimensões lineares do limbo foliar

A estimativa da área foliar pode auxiliar na compreensão de relações de interferência entre plantas daninhas e cultivadas. Com o objetivo de obter uma equação que, por meio de parâmetros lineares dimensionais das folhas, permita a estimativa da área foliar de Sida cordifolia e Sida rhombifolia, estudaram-se as correlações entre área foliar real (Af) e parâmetros dimensionais do limbo foliar, como o comprimento (C) ao longo da nervura principal e a largura máxima (L) perpendicular à nervura principal. Foram analisados 200 limbos foliares de cada espécie, coletados em diferentes agroecossistemas na Universidade Estadual Paulista, campus de Jaboticabal. Os modelos estatísticos utilizados foram linear: Y = a + bx; linear simples: Y = bx; geométrico: Y = ax b; e exponencial: Y = ab x. Todos os modelos analisados podem ser empregados para estimação da área foliar de S. cordifolia e S. rhombifolia. Sugere-se optar pela equação linear simples, envolvendo o produto C*L, considerando-se o coeficiente linear igual a zero, em função da praticidade desta. Desse modo, a estimativa da área foliar de S. cordifolia pode ser obtida pela fórmula Af = 0,7878*(C*L), com coeficiente de determinação de 0,9307, enquanto para S. rhombifolia a estimativa da área foliar pode ser obtida pela fórmula Af = 0,6423*(C*L), com coeficiente de determinação de 0,9711.

Estimativa do número de frutos por amostragem de parte da copa em laranjeiras

Estudou-se um método objetivo para a estimativa do número de frutos em pomares de laranja baseado na contagem dos frutos em ramos de 5 cm de diâmetro. Foram realizados levantamentos em laranjeiras durante três safras, obtendo-se o número de frutos produzidos em um ramo terminal, tomado ao acaso, bem como o número total na árvore. Consideraram-se nove estratos, constituídos pelas cultivares de laranja-doce, Hamlin, Pêra, Natal e Valência (as duas últimas analisadas conjuntamente) e três faixas etárias (três a cinco, seis a 10 e mais que 10 anos de idade). Foram ajustados modelos de regressão linear para o número total de frutos da árvore em função do número de frutos no ramo, obtendo-se coeficientes de determinação variando de 0,79 a 0,94. Com exceção da cultivar Hamlin, verificou-se coincidência entre as curvas das faixas etárias correspondentes. Esses resultados permitem estimar a produção média de frutos em um pomar de laranja, com base em amostragem de ramos com tamanho fixo, com precisão satisfatória, sem o uso de métodos de amostragem mais laboriosos e onerosos.

Equação de regressão linear múltipla para estimativa do erro experimental

Este trabalho teve por objetivo estimar equações de regressão linear múltipla tendo, como variáveis explicativas, as demais características avaliadas em experimento de milho e, como variáveis principais, a diferença mínima significativa em percentagem da média (DMS%) e quadrado médio do erro (QMe), para peso de grãos. Com 610 experimentos conduzidos na Rede de Ensaios Nacionais de Competição de Cultivares de Milho, realizados entre 1986 e 1996 (522 experimentos) e em 1997 (88 experimentos), estimaram-se duas equações de regressão, com os 522 experimentos, validando estas pela análise de regressão simples entre os valores reais e os estimados pelas equações, com os 88 restantes, observando que, para a DMS% a equação não estimava o mesmo valor que a fórmula original e, para o QMe, a equação poderia ser utilizada na estimação. Com o teste de Lilliefors, verificou-se que os valores do QMe aderiam à distribuição normal padrão e foi construída uma tabela de classificação dos valores do QMe, baseada nos valores observados na análise da variância dos experimentos e nos estimados pela equação de regressão.