Bayesian model selection methodology for road safety

Road accidents are a very relevant issue in many countries and macroeconomic models are very frequently applied by academia and administrations to reduce their frequency and consequences. The selection of explanatory variables and response transformation parameter within the Bayesian framework for the selection of the set of explanatory variables a TIM and 3IM (two input and three input models) procedures are proposed. The procedure also uses the DIC and pseudo -R2 goodness of fit criteria. The model to which the methodology is applied is a dynamic regression model with Box-Cox transformation (BCT) for the explanatory variables and autorgressive (AR) structure for the response. The initial set of 22 explanatory variables are identified. The effects of these factors on the fatal accident frequency in Spain, during 2000-2012, are estimated. The dependent variable is constructed considering the stochastic trend component.

Transformed symmetric models

For the first time, we introduce a class of transformed symmetric models to extend the Box and Cox models to more general symmetric models. The new class of models includes all symmetric continuous distributions with a possible non-linear structure for the mean and enables the fitting of a wide range of models to several data types. The proposed methods offer more flexible alternatives to Box-Cox or other existing procedures. We derive a very simple iterative process for fitting these models by maximum likelihood, whereas a direct unconditional maximization would be more difficult. We give simple formulae to estimate the parameter that indexes the transformation of the response variable and the moments of the original dependent variable which generalize previous published results. We discuss inference on the model parameters. The usefulness of the new class of models is illustrated in one application to a real dataset.

Robust automatic mapping algorithms in a network monitoring scenario

Automatically generating maps of a measured variable of interest can be problematic. In this work we focus on the monitoring network context where observations are collected and reported by a network of sensors, and are then transformed into interpolated maps for use in decision making. Using traditional geostatistical methods, estimating the covariance structure of data collected in an emergency situation can be difficult. Variogram determination, whether by method-of-moment estimators or by maximum likelihood, is very sensitive to extreme values. Even when a monitoring network is in a routine mode of operation, sensors can sporadically malfunction and report extreme values. If this extreme data destabilises the model, causing the covariance structure of the observed data to be incorrectly estimated, the generated maps will be of little value, and the uncertainty estimates in particular will be misleading. Marchant and Lark [2007] propose a REML estimator for the covariance, which is shown to work on small data sets with a manual selection of the damping parameter in the robust likelihood. We show how this can be extended to allow treatment of large data sets together with an automated approach to all parameter estimation. The projected process kriging framework of Ingram et al. [2007] is extended to allow the use of robust likelihood functions, including the two component Gaussian and the Huber function. We show how our algorithm is further refined to reduce the computational complexity while at the same time minimising any loss of information. To show the benefits of this method, we use data collected from radiation monitoring networks across Europe. We compare our results to those obtained from traditional kriging methodologies and include comparisons with Box-Cox transformations of the data. We discuss the issue of whether to treat or ignore extreme values, making the distinction between the robust methods which ignore outliers and transformation methods which treat them as part of the (transformed) process. Using a case study, based on an extreme radiological events over a large area, we show how radiation data collected from monitoring networks can be analysed automatically and then used to generate reliable maps to inform decision making. We show the limitations of the methods and discuss potential extensions to remedy these.

A demanda por moeda no Brasil no período 1974-2008: estimativas sob expectativas adaptativas e racionais

O objetivo principal desta dissertação é analisar a demanda por moeda no Brasil no período 1974 a 2008, lembrando que ele inclui sub-períodos de inflação elevada, e baixa, e levando em conta hipóteses alternativas quanto à formação de expectativas. A especificação adotada é a de Tourinho (1995), que generaliza a de Cagan (1956) para permitir uma forma funcional mais flexível e incorporar outras variáveis, além da inflação esperada, como variáveis explicativas. Verifica-se que estas extensões são importantes para modelar a demanda por saldos monetários reais no período aqui considerado. A forma funcional semi-log de Cagan é rejeitada, em favor de uma forma funcional flexível Box-Cox, e os coeficientes da taxa de juros real e da variância da inflação são significativos, mostrando a importância destas variáveis serem inseridas ao modelo. A função estimada para o período completo é comparada com aquelas estimadas para os sub-periodos de inflação alta e moderada, para verificar a estabilidade da formulação adotada. Conclui-se que se pode rejeitar a hipótese de que ela é estável. O modelo de Cagan é generalizado aqui em outra dimensão, considerando mecanismos alternativos de formação de expectativas, que podem ser adaptativas, como no modelo original, ou racionais. A hipótese de que expectativas adaptativas sejam racionais é também considerada. Conclui-se que a imposição da condição de racionalidade ao modelo com expectativas adaptativas não produz alterações importantes nos valores estimados.

辽北农田土壤农药残留特征及其对农产品安全影响

农药对产地环境，特别是对土壤的广泛污染严重威胁农产品安全和人类健康。因此，本文采用建立的除草剂和有机氯农药（OCPs）残留分析方法，开展了辽北地区土壤农药残留特征、阿特拉津和乙草胺田间消解动力学、土壤农药残留对农产品安全影响等方面研究。主要研究结果如下： 1. 分别建立了土壤、大米、蔬菜、玉米中3种除草剂和8种OCPs多残留分析方法。方法检出限介于0.04~1.30 ng•g-1之间；11种农药在0.01 (0.02)~1.0 (2.0) mg•L-1范围内线性良好，相关系数介于0.9963-0.9998之间；平均回收率介于71%-117%之间、相对标准偏差小于14.4%。 2. 阿特拉津和乙草胺在辽北农田土壤普遍残留；丁草胺、六氯苯、狄氏剂和艾氏剂在部分土壤有残留；乙草胺和丁草胺相对其它农药残留较高；阿特拉津、六氯苯、狄氏剂和艾氏剂残留量与相关报道和标准相比较低。除艾氏剂外，检出农药残留量经Box-Cox变换后，均服从正态分布。阿特拉津、乙草胺、丁草胺、六氯苯在不同土壤利用类型之间存在显著差异（P<0.05）。 3. 玉米地土壤中阿特拉津和乙草胺消解动态符合一级反应动力学模式，阿特拉津消解半衰期在12.2~59.8d之间，乙草胺在18.5~54.6d之间。喷施地阿特拉津和乙草胺消解速率约为对照地的2~5倍，且喷施量越大，消解越快。 4. 11种农药在辽北蔬菜、大米、玉米中残留较低，仅阿特拉津、六氯苯、乙草胺和丁草胺在部分农产品中有残留，其在土壤中残留通过蔬菜、大米和玉米给消费者带来的总膳食风险较低。大田试验进一步说明在试验区域喷施4倍最大推荐剂量阿特拉津或乙草胺也不会对玉米安全产生影响。

Genetic heterogeneity of within-family variance of body weight in Atlantic salmon (Salmo salar)

BACKGROUND: Canalization is defined as the stability of a genotype against minor variations in both environment and genetics. Genetic variation in degree of canalization causes heterogeneity of within-family variance. The aims of this study are twofold: (1) quantify genetic heterogeneity of (within-family) residual variance in Atlantic salmon and (2) test whether the observed heterogeneity of (within-family) residual variance can be explained by simple scaling effects. RESULTS: Analysis of body weight in Atlantic salmon using a double hierarchical generalized linear model (DHGLM) revealed substantial heterogeneity of within-family variance. The 95% prediction interval for within-family variance ranged from ~0.4 to 1.2 kg2, implying that the within-family variance of the most extreme high families is expected to be approximately three times larger than the extreme low families. For cross-sectional data, DHGLM with an animal mean sub-model resulted in severe bias, while a corresponding sire-dam model was appropriate. Heterogeneity of variance was not sensitive to Box-Cox transformations of phenotypes, which implies that heterogeneity of variance exists beyond what would be expected from simple scaling effects. CONCLUSIONS: Substantial heterogeneity of within-family variance was found for body weight in Atlantic salmon. A tendency towards higher variance with higher means (scaling effects) was observed, but heterogeneity of within-family variance existed beyond what could be explained by simple scaling effects. For cross-sectional data, using the animal mean sub-model in the DHGLM resulted in biased estimates of variance components, which differed substantially both from a standard linear mean animal model and a sire-dam DHGLM model. Although genetic differences in canalization were observed, selection for increased canalization is difficult, because there is limited individual information for the variance sub-model, especially when based on cross-sectional data. Furthermore, potential macro-environmental changes (diet, climatic region, etc.) may make genetic heterogeneity of variance a less stable trait over time and space.

Testing production functions used in empirical growth studies

We estimate and test two alternative functional forms, which have been used in the growth literature, representing the aggregate production function for a panel of countries: the model of Mankiw, Romer and Weil (Quarterly Journal of Economics, 1992), and a mincerian formulation of schooling-returns to skills. Estimation is performed using instrumental-variable techniques, and both functional forms are confronted using a Box-Cox test, since human capital inputs enter in levels in the mincerian specification and in logs in the extended neoclassical growth model.

Aprimoramento estatístico da regionalização de vazões máximas e médias. Aplicação a bacias hidrográficas do Rio Grande do Sul e Santa Catarina

Os registros de vazões líquidas obtidos das estações fluviométricas são válidos, unicamente, para o local de onde foram coletados ou muito próximo a ele. Na maioria das vezes, a região de influência deste não inclui o local de interesse para o desenvolvimento de projetos de recursos hídricos. Este inconveniente, geralmente, pode ser resolvido através do uso de métodos de regionalização hidrológica. Para determinar os coeficientes da equação de regionalização, o procedimento mais usado consiste em: i) estabelecer uma relação do tipo exponencial multivariada entre a variável dependente (vazão média de longo prazo ou média anual de cheia) e as covariáveis (variáveis climáticas e fisiográficas da bacia hidrográfica); ii) linearizar a equação anterior mediante a transformação logarítmica de ambos os membros; iii) utilizar modelos lineares de regressão para estimar os coeficientes, geralmente, o método dos mínimos quadrados ordinários; e iv) aplicar a transformação inversa para definir a equação. A aplicação deste procedimento implica assumir certas propriedades dos dados (assimetria positiva, registros da mesma extensão e que os mesmos possuem o mesmo período de início e fim, entre outros) que dificilmente podem ser atendidas, prejudicando a verificação das hipóteses nas quais estão baseados os métodos lineares de regressão e, em conseqüência, seu adequado uso, bem como a confiabilidade dos resultados obtidos. Esta pesquisa apresenta um aprimoramento dos métodos de regionalização de vazões geralmente empregados, incluindo-se técnicas que levam em consideração as limitações anteriores. Estas técnicas foram: i) uso da transformada de Box-Cox na linearização da equação exponencial multivariada; ii) determinação dos coeficientes da equação de regionalização usando mínimos quadrados ponderados; e iii) verificação se os resíduos da regressão estão correlacionados ou não. Para o desenvolvimento e verificação da metodologia proposta foram usados somente registros fluviométricos de Bacias Hidrográficas Brasileiras, que drenam suas águas para o Rio Grande do Sul e/ou que estejam localizadas dentro dele. Geograficamente, a área de estudo inclui a totalidade do estado do Rio Grande do Sul e parte de Santa Catarina. As equações de regionalização foram definidas usando dados de vazões médias de longo prazo e média de cheia, para tempo de retorno de 2,33 e 50 anos. Neste último caso, as freqüências foram estimadas através do método dos momentos-L.Comparando os resultados obtidos utilizando o modelo de regionalização hidrológica proposto neste trabalho (transformada de Box-Cox / mínimos quadrados ponderados) junto a seus similares gerados usando a metodologia convencional (transformada logarítmica / mínimos quadrados ordinários) e de modelos intermediários (transformada logarítmica / mínimos quadrados ponderados e transformada de Box-Cox / mínimos quadrados ordinários), os mesmos podem ser considerados satisfatórios, visto que em todas as simulações realizadas o modelo proposto forneceu melhores resultados que aqueles obtidos com os outros modelos, sendo utilizado como padrão de comparação: 1) a qualidade do ajuste, 2) o grau de verificação das hipóteses dos métodos lineares de regressão e 3) os erros na estimativa das descargas, em termos de vazão específica. Nas simulações realizadas usando os modelos intermediários, observou-se que: i) na regionalização de vazões médias, o ganho de considerar a heterogeneidade temporal dos dados é maior do que corrigir a assimetria dos mesmos; ii) quando são usadas séries de descargas máximas, ocorre o efeito contrário, visto que o ganho de corrigir a assimetria das séries é maior do que o efeito da heterogeneidade temporal dos dados. Com relação aos resíduos da regressão, contrariamente ao esperado, os mesmos não sugerem estar correlacionados; isto pode ser conseqüência de utilizar como variável dependente um único registro por estação (vazão média de longo prazo ou média anual de cheia).

Testing production functions used in empirical growth studies

We estimate and test two alternative functional forms representing the aggregate production function for a panel of countries: the extended neoclassical growth model, and a mincerian formulation of schooling-returns to skills. Estimation is performed using instrumentalvariable techniques, and both functional forms are confronted using a Box-Cox test, since human capital inputs enter in levels in the mincerian specification and in logs in the extended neoclassical growth model. Our evidence rejects the extended neoclassical growth model in favor of the mincerian specification, with an estimated capital share of about 42%, a marginal return to education of about 7.5% per year, and an estimated productivity growth of about 1.4% per year. Differences in productivity cannot be disregarded as an explanation of why output per worker varies so much across countries: a variance decomposition exercise shows that productivity alone explains 54% of the variation in output per worker across countries.

Análise Bayesiana de dados composicionais na presença de covariáveis

Pós-graduação em Matematica Aplicada e Computacional - FCT

Box and cox; a romance of real life, in one act.

Slighty cut for acting.

COX-derived prostanoid pathways in gastrointestinal cancer development and progression : novel targets for prevention and intervention

Arachidonic acid metabolism through cyclooxygenase (COX) pathways leads to the generation of biologically active eicosanoids. Eicosanoid expression levels vary during development and progression of gastrointestinal (GI) malignancies. COX-2 is the major COX-isoform responsible for G.I. cancer development/progression. COX-2 expression increases during progression from a normal to cancerous state. Evidence from observational studies has demonstrated that chronic NSAID use reduces the risk of cancer development, while both incidence and risk of death due to G.I. cancers were significantly reduced by daily aspirin intake. A number of randomized controlled trials (APC trial, Prevention of Sporadic Adenomatous Polyps trial, APPROVe trial) have also shown a significant protective effect in patients receiving selective COX-2 inhibitors. However, chronic use of selective COX-2 inhibitors at high doses was associated with increased cardiovascular risk, while NSAIDs have also been associated with increased risk. More recently, downstream effectors of COX-signaling have been investigated in cancer development/progression. PGE 2, which binds to both EP and PPAR receptors, is the major prostanoid implicated in the carcinogenesis of G.I. cancers. The role of TXA 2 in G.I. cancers has also been examined, although further studies are required to uncover its role in carcinogenesis. Other prostanoids investigated include PGD 2 and its metabolite 15d-PGJ2, PGF 1α and PGI 2. Targeting these prostanoids in G.I. cancers has the promise of avoiding cardiovascular toxicity associated with chronic selective COX-2 inhibition, while maintaining anti-tumor reactivity.A progressive sequence from normal to pre-malignant to a malignant state has been identified in G.I. cancers. In this review, we will discuss the role of the COX-derived prostanoids in G.I. cancer development and progression. Targeting these downstream prostanoids for chemoprevention and/or treatment of G.I. cancers will also be discussed. Finally, we will highlight the latest pre-clinical technologies as well as avenues for future investigation in this highly topical research field. © 2011 Elsevier B.V.

An overview of Australian intrans*igence : a systemic Pandora’s Box

Despite the fact that Australia is a socially progressive country and boasts one of the largest Gender Dysphoria Clinics in the Southern Hemisphere, delivering services for almost four decades, Australian Governments fail to arrive at any consensus on the legal and human rights approaches to Trans*people. The subsequent lack of recognition does little more than increase the levels of frustration of and the continued discrimination to Trans*people, including adverse mental health problems, in this country. The purpose of this presentation is to provide an overview of the Australian systems that govern Trans*people and to identify how Trans*identities are manipulated in our Federal system of government; a system which offers little to protect the human rights of Trans*people. In order to contextualise the Australian situation, I commence with a brief description on the layers of government which will include how Australian Trans*people are currently protected under the law in those jurisdictions. I then present some of the impracticalities endured by the transitioning individual (single or married) including change of documentation and legal gender status before, during and after surgical transition for both those born on and off shore. This presentation will also include discussion of legislation that has been described by Trans*advocates as “Gesture”, “Cart before the Horse” and “Harmful”. I will conclude with a way forward by suggesting the development of a coordinated all of government approach in consultation with key stakeholders for “Trans* Friendly Legislation” to improve the human and legal rights, and ultimately the health and wellbeing of Australian Trans*identities.

Polymorphs and hydrates of Etoricoxib, a selective COX-2 inhibitor

The crystal structures of two polymorphs and two polymorphic hemihydrates of Etoricoxib are reported. Etoricoxib is a non-steroidal anti-inflammatory drug (NSAID) that is a selective inhibitor of COX-2. It is used in the treatment of various types of inflammation, pain and fever. Clas et al. have reported four polymorphs (labeled I through IV) and two solvates (hemi-and sesquihydrate) of the API in US patent 6,441,002 (Clas et al, US patent 6,441,002, 2002). However, no crystal structures have been reported for any of these forms. A comparison was made between the PXRD patterns reported in patent `002 and the powder spectra simulated from single crystal data. The two polymorphs characterized here correspond to form I and form IV of the patent. Form II of the patent could not be obtained by us with a variety of experimental conditions. Form III of the patent corresponds to hemihydrate II of this study. Form III is therefore not a polymorph of form I and form IV. What we have termed hemihydrate I in this study is obtained under a wide variety of conditions and it is also the only hemihydrate reported as such in the patent. Because the Etoricoxib molecule contains no conventional hydrogen bond donors, there cannot be any strong hydrogen bonds in the crystal structures of forms I and IV. The packing is accordingly characterized by weak hydrogen bonds of the C-H center dot center dot center dot O=S and C-H center dot center dot center dot N type. Thermal data were collected for form I, form IV and hemihydrate I to shed some light on relative stabilities. PXRD diffractograms show the transformation of form IV to form I at elevated temperature, indicating that form I is more stable than form IV. However, this transformation occurs only in samples of form IV that contain some form I; it does not occur in pure form IV. The formation of the two hemihydrates could follow from the known tendency of an acceptor-rich molecule to crystallize as a hydrate.

data transformation and attribute subset selection: do they help make differences in software failure prediction?

IEEE, IEEE Comp Soc, Tech Council Software Engn