Inference From Aging Information

For many learning tasks the duration of the data collection can be greater than the time scale for changes of the underlying data distribution. The question we ask is how to include the information that data are aging. Ad hoc methods to achieve this include the use of validity windows that prevent the learning machine from making inferences based on old data. This introduces the problem of how to define the size of validity windows. In this brief, a new adaptive Bayesian inspired algorithm is presented for learning drifting concepts. It uses the analogy of validity windows in an adaptive Bayesian way to incorporate changes in the data distribution over time. We apply a theoretical approach based on information geometry to the classification problem and measure its performance in simulations. The uncertainty about the appropriate size of the memory windows is dealt with in a Bayesian manner by integrating over the distribution of the adaptive window size. Thus, the posterior distribution of the weights may develop algebraic tails. The learning algorithm results from tracking the mean and variance of the posterior distribution of the weights. It was found that the algebraic tails of this posterior distribution give the learning algorithm the ability to cope with an evolving environment by permitting the escape from local traps.

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.

Bayesian Analysis for Nonlinear Regression Model Under Skewed Errors, with Application in Growth Curves

We have considered a Bayesian approach for the nonlinear regression model by replacing the normal distribution on the error term by some skewed distributions, which account for both skewness and heavy tails or skewness alone. The type of data considered in this paper concerns repeated measurements taken in time on a set of individuals. Such multiple observations on the same individual generally produce serially correlated outcomes. Thus, additionally, our model does allow for a correlation between observations made from the same individual. We have illustrated the procedure using a data set to study the growth curves of a clinic measurement of a group of pregnant women from an obstetrics clinic in Santiago, Chile. Parameter estimation and prediction were carried out using appropriate posterior simulation schemes based in Markov Chain Monte Carlo methods. Besides the deviance information criterion (DIC) and the conditional predictive ordinate (CPO), we suggest the use of proper scoring rules based on the posterior predictive distribution for comparing models. For our data set, all these criteria chose the skew-t model as the best model for the errors. These DIC and CPO criteria are also validated, for the model proposed here, through a simulation study. As a conclusion of this study, the DIC criterion is not trustful for this kind of complex model.

Generalized Birnbaum-Saunders distributions applied to air pollutant concentration

The generalized Birnbaum-Saunders (GBS) distribution is a new class of positively skewed models with lighter and heavier tails than the traditional Birnbaum-Saunders (BS) distribution, which is largely applied to study lifetimes. However, the theoretical argument and the interesting properties of the GBS model have made its application possible beyond the lifetime analysis. The aim of this paper is to present the GBS distribution as a useful model for describing pollution data and deriving its positive and negative moments. Based on these moments, we develop estimation and goodness-of-fit methods. Also, some properties of the proposed estimators useful for developing asymptotic inference are presented. Finally, an application with real data from Environmental Sciences is given to illustrate the methodology developed. This example shows that the empirical fit of the GBS distribution to the data is very good. Thus, the GBS model is appropriate for describing air pollutant concentration data, which produces better results than the lognormal model when the administrative target is determined for abating air pollution. Copyright (c) 2007 John Wiley & Sons, Ltd.

Surface active ionic liquids: Study of the micellar properties of 1-(1-alkyl)-3-methylimidazolium chlorides and comparison with structurally related surfactants

The impetus for the increasing interest in studying surface active ionic liquids (SAILs; ionic liquids with long-chain ""tails"") is the enormous potential for their applications, e.g., in nanotechnology and biomedicine. The progress in these fields rests on understanding the relationship between surfactant structure and solution properties, hence applications. This need has prompted us to extend our previous study on 1-(1-hexadecyl)-3-methylimidazolium chloride to 1-(1-alkyl)-3-methylimidazolium chlorides, with alkyl chains containing 10, 12, and 14 carbons. In addition to investigating relevant micellar properties, we have compared the solution properties of the imidazolium-based surfactants with: 1-(1-alkyl)pyridinium chlorides, and benzyl (2-acylaminoethyl)dimethylammonium chlorides. The former series carries a heterocyclic ring head-group, but does not possess a hydrogen that is as acidic as H2 of the imidazolium ring. The latter series carries an aromatic ring, a quaternary nitrogen and (a hydrogen-bond forming) amide group. The properties of the imidazolium and pyridinium surfactants were determined in the temperature range from 15 to 75 degrees C. The techniques employed were conductivity, isothermal titration calorimetry, and static light scattering. The results showed the important effects of the interactions in the interfacial region on the micellar properties over the temperature range studied. (C) 2011 Elsevier Inc. All rights reserved.

Do higher moments really matter in portfolio choice?

We present explicit formulas for evaluating the difference between Markowitz weights and those from optimal portfolios, with the same given return, considering either asymmetry or kurtosis. We prove that, whenever the higher moment constraint is not binding, the weights are never the same. If, due to special features of the first and second moments, the difference might be negligible, in quite many cases it will be very significant. An appealing illustration, when the designer wants to incorporate an asset with quite heavy tails, but wants to moderate this effect, further supports the argument.

Comparação de carteiras otimizadas segundo o critério média-variância formadas através de estimativas robustas de risco e retorno

Este trabalho examinou as características de carteiras compostas por ações e otimizadas segundo o critério de média-variância e formadas através de estimativas robustas de risco e retorno. A motivação para isto é a distribuição típica de ativos financeiros (que apresenta outliers e mais curtose que a distribuição normal). Para comparação entre as carteiras, foram consideradas suas propriedades: estabilidade, variabilidade e os índices de Sharpe obtidos pelas mesmas. O resultado geral mostra que estas carteiras obtidas através de estimativas robustas de risco e retorno apresentam melhoras em sua estabilidade e variabilidade, no entanto, esta melhora é insuficiente para diferenciar os índices de Sharpe alcançados pelas mesmas das carteiras obtidas através de método de máxima verossimilhança para estimativas de risco e retorno.

Changes in test scores distribution for students of the fourth grade in Brazil: a relative distribution analysis for the years 1997 to 2005

To assess the quality of school education, much of educational research is concerned with comparisons of test scores means or medians. In this paper, we shift this focus and explore test scores data by addressing some often neglected questions. In the case of Brazil, the mean of test scores in Math for students of the fourth grade has declined approximately 0,2 standard deviation in the late 1990s. But what about changes in the distribution of scores? It is unclear whether the decline was caused by deterioration in student performance in upper and/or lower tails of the distribution. To answer this question, we propose the use of the relative distribution method developed by Handcock and Morris (1999). The advantage of this methodology is that it compares two distributions of test scores data through a single distribution and synthesizes all the differences between them. Moreover, it is possible to decompose the total difference between two distributions in a level effect (changes in median) and shape effect (changes in shape of the distribution). We find that the decline of average-test scores is mainly caused by a worsening in the position of all students throughout the distribution of scores and is not only specific to any quantile of distribution.

Extreme observations and diversification in Latin American emerging equity markets

In this paper, we focus on the tails of the unconditional distribution of Latin American emerging markets stock returns. We explore their implications for portfolio diversification according to the safety tirst principIe, tirst proposed by Roy (1952). We tind that the Latin American emerging markets have signiticantly fatter tails than industrial markets. especially, the lower tail of the distrihution. We consider the implication of the safety tirst principIe for a U .S. investor who creates a diversitied portfolio using Latin American stock markets. We tind that a U.S. investor gains by adding Latin American equity markets to her purely domestic portfolio. For different parameter specitications. we finu a more realistic asset allocation than the one suggested by the Iiterature haseu on the traditional mean-variance framework.

How the full opening of the capital account to highly liquid financial markets led Latin America to two and a half cycles of ‘mania, panic and crash’

Latin America has recently experienced three cycles of capital inflows, the first two ending in major financial crises. The first took place between 1973 and the 1982 ‘debt-crisis’. The second took place between the 1989 ‘Brady bonds’ agreement (and the beginning of the economic reforms and financial liberalisation that followed) and the Argentinian 2001/2002 crisis, and ended up with four major crises (as well as the 1997 one in East Asia) — Mexico (1994), Brazil (1999), and two in Argentina (1995 and 2001/2). Finally, the third inflow-cycle began in 2003 as soon as international financial markets felt reassured by the surprisingly neo-liberal orientation of President Lula’s government; this cycle intensified in 2004 with the beginning of a (purely speculative) commodity price-boom, and actually strengthened after a brief interlude following the 2008 global financial crash — and at the time of writing (mid-2011) this cycle is still unfolding, although already showing considerable signs of distress. The main aim of this paper is to analyse the financial crises resulting from this second cycle (both in LA and in East Asia) from the perspective of Keynesian/ Minskyian/ Kindlebergian financial economics. I will attempt to show that no matter how diversely these newly financially liberalised Developing Countries tried to deal with the absorption problem created by the subsequent surges of inflow (and they did follow different routes), they invariably ended up in a major crisis. As a result (and despite the insistence of mainstream analysis), these financial crises took place mostly due to factors that were intrinsic (or inherent) to the workings of over-liquid and under-regulated financial markets — and as such, they were both fully deserved and fairly predictable. Furthermore, these crises point not just to major market failures, but to a systemic market failure: evidence suggests that these crises were the spontaneous outcome of actions by utility-maximising agents, freely operating in friendly (‘light-touch’) regulated, over-liquid financial markets. That is, these crises are clear examples that financial markets can be driven by buyers who take little notice of underlying values — i.e., by investors who have incentives to interpret information in a biased fashion in a systematic way. Thus, ‘fat tails’ also occurred because under these circumstances there is a high likelihood of self-made disastrous events. In other words, markets are not always right — indeed, in the case of financial markets they can be seriously wrong as a whole. Also, as the recent collapse of ‘MF Global’ indicates, the capacity of ‘utility-maximising’ agents operating in (excessively) ‘friendly-regulated’ and over-liquid financial market to learn from previous mistakes seems rather limited.

How the full opening of the capital account to highly liquid financial markets led Latin America to two and a half cycles of ‘mania, panic and crash’

Latin America has recently experienced three cycles of capital inflows, the first two ending in major financial crises. The first took place between 1973 and the 1982 ‘debt-crisis’. The second took place between the 1989 ‘Brady bonds’ agreement (and the beginning of the economic reforms and financial liberalisation that followed) and the Argentinian 2001/2002 crisis, and ended up with four major crises (as well as the 1997 one in East Asia) — Mexico (1994), Brazil (1999), and two in Argentina (1995 and 2001/2). Finally, the third inflow-cycle began in 2003 as soon as international financial markets felt reassured by the surprisingly neo-liberal orientation of President Lula’s government; this cycle intensified in 2004 with the beginning of a (purely speculative) commodity price-boom, and actually strengthened after a brief interlude following the 2008 global financial crash — and at the time of writing (mid-2011) this cycle is still unfolding, although already showing considerable signs of distress. The main aim of this paper is to analyse the financial crises resulting from this second cycle (both in LA and in East Asia) from the perspective of Keynesian/ Minskyian/ Kindlebergian financial economics. I will attempt to show that no matter how diversely these newly financially liberalised Developing Countries tried to deal with the absorption problem created by the subsequent surges of inflow (and they did follow different routes), they invariably ended up in a major crisis. As a result (and despite the insistence of mainstream analysis), these financial crises took place mostly due to factors that were intrinsic (or inherent) to the workings of over-liquid and under-regulated financial markets — and as such, they were both fully deserved and fairly predictable. Furthermore, these crises point not just to major market failures, but to a systemic market failure: evidence suggests that these crises were the spontaneous outcome of actions by utility-maximising agents, freely operating in friendly (light-touched) regulated, over-liquid financial markets. That is, these crises are clear examples that financial markets can be driven by buyers who take little notice of underlying values — investors have incentives to interpret information in a biased fashion in a systematic way. ‘Fat tails’ also occurred because under these circumstances there is a high likelihood of self-made disastrous events. In other words, markets are not always right — indeed, in the case of financial markets they can be seriously wrong as a whole. Also, as the recent collapse of ‘MF Global’ indicates, the capacity of ‘utility-maximising’ agents operating in unregulated and over-liquid financial market to learn from previous mistakes seems rather limited.

Whole-genome analysis for backfat thickness in a tropically adapted, composite cattle breed from Brazil

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Estudo morfométrico comparativo de populações de Rotylenchulus reniformis (Nemata: Rotylenchulinae) do Brasil

Quarenta e oito populações de Rotylenchulus reniformis foram recuperadas de amostras de solo e raízes de diferentes culturas e inoculadas em diferentes plantas hospedeiras, mantidas em microparcelas no Departamento de Fitossanidade da UNESP/FCAV, Campus de Jaboticabal, São Paulo. Os sintomas da doença causada pelo nematóide em algodoeiro (Gossypium hirsutum), no campo, foram documentados, bem como o hábito de parasitismo do nematóide em raízes de algodão e de mamoeiro (Carica papaya), utilizando-se de coloração in situ do nematóide com fuccina ácida. Efetuou-se a comparação morfológica de todas as populações, ao microscópio óptico composto, em montagens temporárias e, de algumas, ao microscópio eletrônico de varredura. Para as observações ao microscópio eletrônico de varredura, fêmeas adultas presas às raízes foram fixadas em glutaraldeído e pós-fixadas em tetróxido de ósmio, desidratadas em álcool etílico, secas em secador de ponto crítico, montadas, recobertas com 35 nm de ouro, observadas e elétromicrografadas em 15 kV. Os dados obtidos confirmam que R. reniformis é a única espécie do gênero distribuída nos agroecossistemas brasileiros e que a amplitude de variação de caracteres morfométricos em populações brasileiras desse nematóide, tais como comprimento do estilete, V % e forma da cauda, é maior que em populações da mesma espécie de outras regiões do mundo. Foram ilustradas fêmeas jovens de R. reniformis com a cauda bifurcada, e esse detalhe da morfologia do nematóide ainda não havia sido relatado.

Remoção conjugada de metais e óleo de água produzida

Petroleum can be associated or not with natural gas, but in both cases water is always present in its formation. The presence of water causes several problems, such as the difficulty of removing the petroleum from the reservoir rock and the formation of waterin-oil and oil-in-water emulsions. The produced water causes environmental problems, which should be solved to reduce the effect of petroleum industry in the environment. The main objective of this work is to remove simultaneously from the produced water the dispersed petroleum and dissolved metals. The process is made possible through the use of anionic surfactants that with its hydrophilic heads interacts with ionized metals and with its lipophilic tails interacts with the oil. The studied metals were: calcium, magnesium, barium, and cadmium. The surfactants used in this research were derived from: soy oil, sunflower oil, coconut oil, and a soap obtained from a mixture of 5wt.% coconut oil and 95wt.% animal fat. It was used a sample of produced water from Terminal de São Sebastião, São Paulo. As the concentration of the studied metals in produced water presented values close to 300 mg/L, it was decided to use this concentration as reference for the development of this research. Molecular absorption and atomic absorption spectroscopy were used to determine petroleum and metals concentrations in the water sample, respectively. A constant pressure filtration system was used to promote the separation of solid and liquid phases. To represent the behavior of the studied systems it was developed an equilibrium model and a mathematical one. The obtained results showed that all used surfactants presented similar behavior with relation to metals extraction, being selected the surfactant derived from soy oil for this purpose. The values of the partition coefficients between the solid and liquid phases " D " for the studied metals varied from 0.2 to 1.1, while the coefficients for equilibrium model " K " varied from 0.0002 and 0.0009. The removal percentile for oil with all metals associated was near 100%, showing the efficiency of the process

Loss of nuclear poly(A)-binding protein 1 causes defects in myogenesis and mRNA biogenesis

The nuclear poly(A)-binding protein 1 (PABPN1) is a ubiquitously expressed protein that plays a critical role in polyadenylation. Short expansions of the polyalanine tract in the N-terminus of PABPN1 lead to oculopharyngeal muscular dystrophy (OPMD), which is an adult onset disease characterized by eyelid drooping, difficulty in swallowing and weakness in the proximal limb muscles. Although significant data from in vitro biochemical assays define the function of PABPN1 in control of poly(A) tail length, little is known about the role of PABPN1 in mammalian cells. To assess the function of PABPN1 in mammalian cells and specifically in cells affected in OPMD, we examined the effects of PABPN1 depletion using siRNA in primary mouse myoblasts from extraocular, pharyngeal and limb muscles. PABPN1 knockdown significantly decreased cell proliferation and myoblast differentiation during myogenesis in vitro. At the molecular level, PABPN1 depletion in myoblasts led to a shortening of mRNA poly(A) tails, demonstrating the cellular function of PABPN1 in polyadenylation control in a mammalian cell. In addition, PABPN1 depletion caused nuclear accumulation of poly(A) RNA, revealing that PABPN1 is required for proper poly(A) RNA export from the nucleus. Together, these experiments demonstrate that PABPN1 plays an essential role in myoblast proliferation and differentiation, suggesting that it is required for muscle regeneration and maintenance in vivo.