A comparison of the alr and ilr transformations for kernel density estimation of compositional data

In a seminal paper, Aitchison and Lauder (1985) introduced classical kernel densityestimation techniques in the context of compositional data analysis. Indeed, they gavetwo options for the choice of the kernel to be used in the kernel estimator. One ofthese kernels is based on the use the alr transformation on the simplex SD jointly withthe normal distribution on RD-1. However, these authors themselves recognized thatthis method has some deficiencies. A method for overcoming these dificulties based onrecent developments for compositional data analysis and multivariate kernel estimationtheory, combining the ilr transformation with the use of the normal density with a fullbandwidth matrix, was recently proposed in Martín-Fernández, Chacón and Mateu-Figueras (2006). Here we present an extensive simulation study that compares bothmethods in practice, thus exploring the finite-sample behaviour of both estimators

Adaptively weighted maximum likelihood estimation of discrete distributions

SummaryDiscrete data arise in various research fields, typically when the observations are count data.I propose a robust and efficient parametric procedure for estimation of discrete distributions. The estimation is done in two phases. First, a very robust, but possibly inefficient, estimate of the model parameters is computed and used to indentify outliers. Then the outliers are either removed from the sample or given low weights, and a weighted maximum likelihood estimate (WML) is computed.The weights are determined via an adaptive process such that if the data follow the model, then asymptotically no observation is downweighted.I prove that the final estimator inherits the breakdown point of the initial one, and that its influence function at the model is the same as the influence function of the maximum likelihood estimator, which strongly suggests that it is asymptotically fully efficient.The initial estimator is a minimum disparity estimator (MDE). MDEs can be shown to have full asymptotic efficiency, and some MDEs have very high breakdown points and very low bias under contamination. Several initial estimators are considered, and the performances of the WMLs based on each of them are studied.It results that in a great variety of situations the WML substantially improves the initial estimator, both in terms of finite sample mean square error and in terms of bias under contamination. Besides, the performances of the WML are rather stable under a change of the MDE even if the MDEs have very different behaviors.Two examples of application of the WML to real data are considered. In both of them, the necessity for a robust estimator is clear: the maximum likelihood estimator is badly corrupted by the presence of a few outliers.This procedure is particularly natural in the discrete distribution setting, but could be extended to the continuous case, for which a possible procedure is sketched.RésuméLes données discrètes sont présentes dans différents domaines de recherche, en particulier lorsque les observations sont des comptages.Je propose une méthode paramétrique robuste et efficace pour l'estimation de distributions discrètes. L'estimation est faite en deux phases. Tout d'abord, un estimateur très robuste des paramètres du modèle est calculé, et utilisé pour la détection des données aberrantes (outliers). Cet estimateur n'est pas nécessairement efficace. Ensuite, soit les outliers sont retirés de l'échantillon, soit des faibles poids leur sont attribués, et un estimateur du maximum de vraisemblance pondéré (WML) est calculé.Les poids sont déterminés via un processus adaptif, tel qu'asymptotiquement, si les données suivent le modèle, aucune observation n'est dépondérée.Je prouve que le point de rupture de l'estimateur final est au moins aussi élevé que celui de l'estimateur initial, et que sa fonction d'influence au modèle est la même que celle du maximum de vraisemblance, ce qui suggère que cet estimateur est pleinement efficace asymptotiquement.L'estimateur initial est un estimateur de disparité minimale (MDE). Les MDE sont asymptotiquement pleinement efficaces, et certains d'entre eux ont un point de rupture très élevé et un très faible biais sous contamination. J'étudie les performances du WML basé sur différents MDEs.Le résultat est que dans une grande variété de situations le WML améliore largement les performances de l'estimateur initial, autant en terme du carré moyen de l'erreur que du biais sous contamination. De plus, les performances du WML restent assez stables lorsqu'on change l'estimateur initial, même si les différents MDEs ont des comportements très différents.Je considère deux exemples d'application du WML à des données réelles, où la nécessité d'un estimateur robuste est manifeste : l'estimateur du maximum de vraisemblance est fortement corrompu par la présence de quelques outliers.La méthode proposée est particulièrement naturelle dans le cadre des distributions discrètes, mais pourrait être étendue au cas continu.

Determine Initial Cause for Current Premature Portland Cement Concrete Pavement Deterioration Final Report, 2000

A detailed investigation has been conducted on core samples taken from 17 portland cement concrete pavements located in Iowa. The goal of the investigation was to help to clarify the root cause of the premature deterioration problem that has become evident since the early 1990s. Laboratory experiments were also conducted to evaluate how cement composition, mixing time, and admixtures could have influenced the occurrence of premature deterioration. The cements used in this study were selected in an attempt to cover the main compositional parameters pertinent to the construction industry in Iowa. The hardened air content determinations conducted during this study indicated that the pavements that exhibited premature deterioration often contained poor to marginal entrained-air void systems. In addition, petrographic studies indicated that sometimes the entrained-air void system had been marginal after mixing and placement of the pavement slab, while in other instances a marginal to adequate entrained-air void system had been filled with ettringite. The filling was most probably accelerated because of shrinkage cracking at the surface of the concrete pavements. The results of this study suggest that the durability—more sciecifically, the frost resistance—of the concrete pavements should be less than anticipated during the design stage of the pavements. Construction practices played a significant role in the premature deterioration problem. The pavements that exhibited premature distress also exhibited features that suggested poor mixing and poor control of aggregate grading. Segregation was very common in the cores extracted from the pavements that exhibited premature distress. This suggests that the vibrators on the paver were used to overcome a workability problem. Entrained-air voids formed in concrete mixtures experiencing these types of problems normally tend to be extremely coarse, and hence they can easily be lost during the paving process. This tends to leave the pavement with a low air content and a poor distribution of air voids. All of these features were consistent with a premature stiffening problem that drastically influenced the ability of the contractor to place the concrete mixture. Laboratory studies conducted during this project indicated that most premature stiffening problems can be directly attributed to the portland cement used on the project. The admixtures (class C fly ash and water reducer) tended to have only a minor influence on the premature stiffening problem when they were used at the dosage rates described in this study.

Sustainable Development and Concrete Technology, May 2004

For a variety of reasons, the concrete construction industry is not sustainable. First, it consumes huge quantities of virgin materials. Second, the principal binder in concrete is portland cement, the production of which is a major contributor to greenhouse gas emissions that are implicated in global warming and climate change. Third, many concrete structures suffer from lack of durability which has an adverse effect on the resource productivity of the industry. Because the high-volume fly ash concrete system addresses all three sustainability issues, its adoption will enable the concrete construction industry to become more sustainable. In this paper, a brief review is presented of the theory and construction practice with concrete mixtures containing more than 50% fly ash by mass of the cementitious material. Mechanisms are discussed by which the incorporation of high volume of fly ash in concrete reduces the water demand, improves the workability, minimizes cracking due to thermal and drying shrinkage, and enhances durability to reinforcement corrosion, sulfate attack, and alkali-silica expansion. For countries like China and India, this technology can play an important role in meeting the huge demand for infrastructure in a sustainable manner.

Is there a wage penalty for horizontal and vertical mismatch?

This paper studies how the horizontal and vertical mismatches in the labor market affect wage. We do so by taking into account that by choosing a job, wage and mismatches are simultaneously determined. The Seemingly Unrelated Equations model also allows us to control for any omitted variable that could cause biased estimators. We use REFLEX data for Spain. Results reveal that in most cases being horizontally matched has a wage premium and being over-educated does not affect wage. Results suggest that the modeling strategy successfully accounts for some omitted variable that affects simultaneously the probability of being horizontally matched and the wage. This could explain the existence of a wage penalty for over-educated workers when the omitted variable issue is not dealt with.

Fate of rubrospinal neurons after unilateral section of the cervical spinal cord in adult macaque monkeys: effects of an antibody treatment neutralizing Nogo-A.

The present study describes in primates the effects of a spinal cord injury on the number and size of the neurons in the magnocellular part of the red nucleus (RNm), the origin of the rubrospinal tract, and evaluates whether a neutralization of Nogo-A reduces the lesioned-induced degenerative processes observed in RNm. Two groups of monkeys were subjected to unilateral section of the spinal cord affecting the rubrospinal tract; one group was subsequently treated with an antibody neutralizing Nogo-A; the second group received a control antibody. Intact animals were also included in the study. Counting neurons stained with a monoclonal antibody recognizing non-phosphorylated epitopes on neurofilaments (SMI-32) indicated that their number in the contralesional RNm was consistently inferior to that in the ipsilesional RNm, in a proportion amounting up to 35%. The lesion also induced shrinkage of the soma of the neurons detected in the contralesional RNm. Infusing an anti-Nogo-A antibody at the site of the lesion did not increase the proportion of SMI-32 positive rubrospinal neurons in the contralesional RNm nor prevent shrinkage.

BME-based uncertainty assessment of the Chernobyl fallout

The vast territories that have been radioactively contaminated during the 1986 Chernobyl accident provide a substantial data set of radioactive monitoring data, which can be used for the verification and testing of the different spatial estimation (prediction) methods involved in risk assessment studies. Using the Chernobyl data set for such a purpose is motivated by its heterogeneous spatial structure (the data are characterized by large-scale correlations, short-scale variability, spotty features, etc.). The present work is concerned with the application of the Bayesian Maximum Entropy (BME) method to estimate the extent and the magnitude of the radioactive soil contamination by 137Cs due to the Chernobyl fallout. The powerful BME method allows rigorous incorporation of a wide variety of knowledge bases into the spatial estimation procedure leading to informative contamination maps. Exact measurements (?hard? data) are combined with secondary information on local uncertainties (treated as ?soft? data) to generate science-based uncertainty assessment of soil contamination estimates at unsampled locations. BME describes uncertainty in terms of the posterior probability distributions generated across space, whereas no assumption about the underlying distribution is made and non-linear estimators are automatically incorporated. Traditional estimation variances based on the assumption of an underlying Gaussian distribution (analogous, e.g., to the kriging variance) can be derived as a special case of the BME uncertainty analysis. The BME estimates obtained using hard and soft data are compared with the BME estimates obtained using only hard data. The comparison involves both the accuracy of the estimation maps using the exact data and the assessment of the associated uncertainty using repeated measurements. Furthermore, a comparison of the spatial estimation accuracy obtained by the two methods was carried out using a validation data set of hard data. Finally, a separate uncertainty analysis was conducted that evaluated the ability of the posterior probabilities to reproduce the distribution of the raw repeated measurements available in certain populated sites. The analysis provides an illustration of the improvement in mapping accuracy obtained by adding soft data to the existing hard data and, in general, demonstrates that the BME method performs well both in terms of estimation accuracy as well as in terms estimation error assessment, which are both useful features for the Chernobyl fallout study.

Adverse effects of intraperitoneal onlay mesh used for incisional hernia repair

Context: In the past 50 years, the use of prosthetic mesh in surgery has dramatically¦changed the management of primary, as well as incisional hernias. Currently, there¦are a large number of different mesh brands and no consensus on the best material,¦nor the best mesh implantation technique to use. The purpose of this study is to¦illustrate the adverse effects of intraperitoneal onlay mesh used for incisional¦hernia repair encountered in patients treated at CHUV for complications after¦incisional hernia repair.¦Materials & Methods: This work is an observational retrospective study. A PubMed¦search and a systematic review of literature were performed. Thereafter, the medical¦records of 22 patients who presented with pain, abdominal discomfort, ileus, fistula,¦abscess, seroma, mesh infection or recurrent incisional hernia after a laparoscopic or¦open repair with intra-abdominal mesh were reviewed.¦Results: Twenty-two persons were reoperated for complications after incisional¦hernia repair with a prosthetic mesh. Ten were male and twelve female, with a¦median age of 58,6 years (range 24-82). Mesh placement was performed by a¦laparoscopic approach in nine patients and by open approach in thirteen others.¦Eight different mesh brands were found (Ultrapro®, Mersilene®, Parietex Composite®,¦Proceed®, DynaMesh®, Gore® DualMesh®, Permacol®, Titanium Metals UK Ltd®).¦Mean time from implantation and reoperation for complication was 34.2 months¦(range 1-147). In our sample of 22 patients, 21 (96%) presented mesh adhesion and¦15 (68%) presented hernia recurrence. Others complications like mesh shrinkage,¦mesh migration, nerve entrapment, seroma, fistula and abscess were also evaluated.¦Conclusion: The majority of articles deal with complications induced by¦intraperitoneal prosthetic mesh, but the effectiveness of mesh has been studied¦mostly on experimental models. Actually and as shown in the present study,¦intraperitoneal mesh placement was associated with severe complications witch may¦potentially be life threatening. In our opinion, intraperitoneal mesh placement should¦only be reserved in exceptional situations, when the modified Rives-Stoppa could not¦be achieved and when tissues covering the mesh are insufficient.

Fauna de Muscidae (Diptera) em três localidades do Estado do Paraná, Brasil, capturada por armadilha Malaise

During two years, from August 1986 to July 1988, the entomofauna of some preserved areas of Parana State, southern Brazil, was sampled in a project called "Levantamento da Fauna Entomológica no Estado do Paraná (PROFAUPAR)". Specimens of Muscidae (Diptera) were sorted out from the material collected using Malaise traps in three of the eight sites sampled (Colombo, Ponta Grossa and Guarapuava) in the first year (August 1986 to July 1987). A total of 7,014 specimens of Muscidae was captured and 91 species were identified. Neodexiopsis flavipalpis Albuquerque was the most abundant species in Ponta Grossa (672 specimens) and in Guarapuava (332 specimens). For Colombo, the most abundant species was Neodexiopsis vulgaris Couri & Albuquerque (172 specimens). The highest richness of species and abundance were observed in Ponta Grossa (77 and 3,559 respectively). The total number of specimens and means values of capture were analyzed. Indices of diversity and evenness were used to discuss richness and dominance of species in each locality. Besides using ecological indices, species richness estimators were also used.

Option pricing under stochastic volatility and stochastic interest rate in the Spanish case

Among the underlying assumptions of the Black-Scholes option pricingmodel, those of a fixed volatility of the underlying asset and of aconstantshort-term riskless interest rate, cause the largest empirical biases. Onlyrecently has attention been paid to the simultaneous effects of thestochasticnature of both variables on the pricing of options. This paper has tried toestimate the effects of a stochastic volatility and a stochastic interestrate inthe Spanish option market. A discrete approach was used. Symmetricand asymmetricGARCH models were tried. The presence of in-the-mean and seasonalityeffectswas allowed. The stochastic processes of the MIBOR90, a Spanishshort-terminterest rate, from March 19, 1990 to May 31, 1994 and of the volatilityofthe returns of the most important Spanish stock index (IBEX-35) fromOctober1, 1987 to January 20, 1994, were estimated. These estimators wereused onpricing Call options on the stock index, from November 30, 1993 to May30, 1994.Hull-White and Amin-Ng pricing formulas were used. These prices werecomparedwith actual prices and with those derived from the Black-Scholesformula,trying to detect the biases reported previously in the literature. Whereasthe conditional variance of the MIBOR90 interest rate seemed to be freeofARCH effects, an asymmetric GARCH with in-the-mean and seasonalityeffectsand some evidence of persistence in variance (IEGARCH(1,2)-M-S) wasfoundto be the model that best represent the behavior of the stochasticvolatilityof the IBEX-35 stock returns. All the biases reported previously in theliterature were found. All the formulas overpriced the options inNear-the-Moneycase and underpriced the options otherwise. Furthermore, in most optiontrading, Black-Scholes overpriced the options and, because of thetime-to-maturityeffect, implied volatility computed from the Black-Scholes formula,underestimatedthe actual volatility.

Speed of convergence of recursive least squares learning with ARMA perceptions

This paper fills a gap in the existing literature on least squareslearning in linear rational expectations models by studying a setup inwhich agents learn by fitting ARMA models to a subset of the statevariables. This is a natural specification in models with privateinformation because in the presence of hidden state variables, agentshave an incentive to condition forecasts on the infinite past recordsof observables. We study a particular setting in which it sufficesfor agents to fit a first order ARMA process, which preserves thetractability of a finite dimensional parameterization, while permittingconditioning on the infinite past record. We describe how previousresults (Marcet and Sargent [1989a, 1989b] can be adapted to handlethe convergence of estimators of an ARMA process in our self--referentialenvironment. We also study ``rates'' of convergence analytically and viacomputer simulation.

Honey, I shrunk the sample covariance matrix

The central message of this paper is that nobody should be using the samplecovariance matrix for the purpose of portfolio optimization. It containsestimation error of the kind most likely to perturb a mean-varianceoptimizer. In its place, we suggest using the matrix obtained from thesample covariance matrix through a transformation called shrinkage. Thistends to pull the most extreme coefficients towards more central values,thereby systematically reducing estimation error where it matters most.Statistically, the challenge is to know the optimal shrinkage intensity,and we give the formula for that. Without changing any other step in theportfolio optimization process, we show on actual stock market data thatshrinkage reduces tracking error relative to a benchmark index, andsubstantially increases the realized information ratio of the activeportfolio manager.

Measuring the stability of histogram appearance when the anchor position is changed

Although the histogram is the most widely used density estimator, itis well--known that the appearance of a constructed histogram for a given binwidth can change markedly for different choices of anchor position. In thispaper we construct a stability index $G$ that assesses the potential changesin the appearance of histograms for a given data set and bin width as theanchor position changes. If a particular bin width choice leads to an unstableappearance, the arbitrary choice of any one anchor position is dangerous, anda different bin width should be considered. The index is based on the statisticalroughness of the histogram estimate. We show via Monte Carlo simulation thatdensities with more structure are more likely to lead to histograms withunstable appearance. In addition, ignoring the precision to which the datavalues are provided when choosing the bin width leads to instability. We provideseveral real data examples to illustrate the properties of $G$. Applicationsto other binned density estimators are also discussed.

Weighted Kernel regression

In the fixed design regression model, additional weights areconsidered for the Nadaraya--Watson and Gasser--M\"uller kernel estimators.We study their asymptotic behavior and the relationships between new andclassical estimators. For a simple family of weights, and considering theIMSE as global loss criterion, we show some possible theoretical advantages.An empirical study illustrates the performance of the weighted estimatorsin finite samples.

Drosophilids (Diptera) from an Atlantic Forest Area in Santa Catarina, Southern Brazil

The present work aims at knowing the faunal composition of drosophilids in forest areas of southern Brazil. Besides, estimation of species richness for this fauna is briefly discussed. The sampling were carried out in three well-preserved areas of the Atlantic Rain Forest in the State of Santa Catarina. In this study, 136,931 specimens were captured and 96.6% of them were identified in the specific level. The observed species richness (153 species) is the largest that has been registered in faunal inventories conducted in Brazil. Sixty-three of the captured species did not fit to the available descriptions, and we believe that most of them are non-described species. The incidence-based estimators tended to give rise to the largest richness estimates while the abundance based give rise to the smallest ones. Such estimators suggest the presence from 172.28 to 220.65 species in the studied area. Based on these values, from 69.35 to 88.81% of the expected species richness were sampled. We suggest that the large richness recorded in this study is a consequence of the large sampling effort, the capture method, recent advances in the taxonomy of drosophilids, the high preservation level and the large extension of the sampled fragment and the high complexity of the Atlantic Rain forest. Finally, our data set suggest that the employment of estimators of richness for drosophilid assemblages is useful but it requires caution.