Exact Search Directions for Optimization of Linear and Nonlinear Models Based on Generalized Orthonormal Functions

A novel technique for selecting the poles of orthonormal basis functions (OBF) in Volterra models of any order is presented. It is well-known that the usual large number of parameters required to describe the Volterra kernels can be significantly reduced by representing each kernel using an appropriate basis of orthonormal functions. Such a representation results in the so-called OBF Volterra model, which has a Wiener structure consisting of a linear dynamic generated by the orthonormal basis followed by a nonlinear static mapping given by the Volterra polynomial series. Aiming at optimizing the poles that fully parameterize the orthonormal bases, the exact gradients of the outputs of the orthonormal filters with respect to their poles are computed analytically by using a back-propagation-through-time technique. The expressions relative to the Kautz basis and to generalized orthonormal bases of functions (GOBF) are addressed; the ones related to the Laguerre basis follow straightforwardly as a particular case. The main innovation here is that the dynamic nature of the OBF filters is fully considered in the gradient computations. These gradients provide exact search directions for optimizing the poles of a given orthonormal basis. Such search directions can, in turn, be used as part of an optimization procedure to locate the minimum of a cost-function that takes into account the error of estimation of the system output. The Levenberg-Marquardt algorithm is adopted here as the optimization procedure. Unlike previous related work, the proposed approach relies solely on input-output data measured from the system to be modeled, i.e., no information about the Volterra kernels is required. Examples are presented to illustrate the application of this approach to the modeling of dynamic systems, including a real magnetic levitation system with nonlinear oscillatory behavior.

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.

Robust linear mixed models with skew-normal independent distributions from a Bayesian perspective

Linear mixed models were developed to handle clustered data and have been a topic of increasing interest in statistics for the past 50 years. Generally. the normality (or symmetry) of the random effects is a common assumption in linear mixed models but it may, sometimes, be unrealistic, obscuring important features of among-subjects variation. In this article, we utilize skew-normal/independent distributions as a tool for robust modeling of linear mixed models under a Bayesian paradigm. The skew-normal/independent distributions is an attractive class of asymmetric heavy-tailed distributions that includes the skew-normal distribution, skew-t, skew-slash and the skew-contaminated normal distributions as special cases, providing an appealing robust alternative to the routine use of symmetric distributions in this type of models. The methods developed are illustrated using a real data set from Framingham cholesterol study. (C) 2009 Elsevier B.V. All rights reserved.

A linear optimization approach to the combined production planning model

Two fundamental processes usually arise in the production planning of many industries. The first one consists of deciding how many final products of each type have to be produced in each period of a planning horizon, the well-known lot sizing problem. The other process consists of cutting raw materials in stock in order to produce smaller parts used in the assembly of final products, the well-studied cutting stock problem. In this paper the decision variables of these two problems are dependent of each other in order to obtain a global optimum solution. Setups that are typically present in lot sizing problems are relaxed together with integer frequencies of cutting patterns in the cutting problem. Therefore, a large scale linear optimizations problem arises, which is exactly solved by a column generated technique. It is worth noting that this new combined problem still takes the trade-off between storage costs (for final products and the parts) and trim losses (in the cutting process). We present some sets of computational tests, analyzed over three different scenarios. These results show that, by combining the problems and using an exact method, it is possible to obtain significant gains when compared to the usual industrial practice, which solve them in sequence. (C) 2010 The Franklin Institute. Published by Elsevier Ltd. All rights reserved.

An integer linear programming approach for approximate string comparison

We introduce a problem called maximum common characters in blocks (MCCB), which arises in applications of approximate string comparison, particularly in the unification of possibly erroneous textual data coming from different sources. We show that this problem is NP-complete, but can nevertheless be solved satisfactorily using integer linear programming for instances of practical interest. Two integer linear formulations are proposed and compared in terms of their linear relaxations. We also compare the results of the approximate matching with other known measures such as the Levenshtein (edit) distance. (C) 2008 Elsevier B.V. All rights reserved.

Transformed generalized linear models

The estimation of data transformation is very useful to yield response variables satisfying closely a normal linear model, Generalized linear models enable the fitting of models to a wide range of data types. These models are based on exponential dispersion models. We propose a new class of transformed generalized linear models to extend the Box and Cox models and the generalized linear models. We use the generalized linear model framework to fit these models and discuss maximum likelihood estimation and inference. We give a simple formula to estimate the parameter that index the transformation of the response variable for a subclass of models. We also give a simple formula to estimate the rth moment of the original dependent variable. We explore the possibility of using these models to time series data to extend the generalized autoregressive moving average models discussed by Benjamin er al. [Generalized autoregressive moving average models. J. Amer. Statist. Assoc. 98, 214-223]. The usefulness of these models is illustrated in a Simulation study and in applications to three real data sets. (C) 2009 Elsevier B.V. All rights reserved.

Genetic Algorithm for the Determination of Linear Viscoelastic Relaxation Spectrum from Experimental Data

Conventional procedures employed in the modeling of viscoelastic properties of polymer rely on the determination of the polymer`s discrete relaxation spectrum from experimentally obtained data. In the past decades, several analytical regression techniques have been proposed to determine an explicit equation which describes the measured spectra. With a diverse approach, the procedure herein introduced constitutes a simulation-based computational optimization technique based on non-deterministic search method arisen from the field of evolutionary computation. Instead of comparing numerical results, this purpose of this paper is to highlight some Subtle differences between both strategies and focus on what properties of the exploited technique emerge as new possibilities for the field, In oder to illustrate this, essayed cases show how the employed technique can outperform conventional approaches in terms of fitting quality. Moreover, in some instances, it produces equivalent results With much fewer fitting parameters, which is convenient for computational simulation applications. I-lie problem formulation and the rationale of the highlighted method are herein discussed and constitute the main intended contribution. (C) 2009 Wiley Periodicals, Inc. J Appl Polym Sci 113: 122-135, 2009

Large Changes of Static Electric Properties Induced by Hydrogen Bonding: An ab Initio Study of Linear HCN Oligomers

We report the partitioning of the interaction-induced static electronic dipole (hyper)polarizabilities for linear hydrogen cyanide complexes into contributions arising from various interaction energy terms. We analyzed the nonadditivities of the studied properties and used these data to predict the electric properties of an infinite chain. The interaction-induced static electric dipole properties and their nonadditivities were analyzed using an approach based on numerical differentiation of the interaction energy components estimated in an external electric field. These were obtained using the hybrid variational-perturbational interaction energy decomposition scheme, augmented with coupled-cluster calculations, with singles, doubles, and noniterative triples. Our results indicate that the interaction-induced dipole moments and polarizabilities are primarily electrostatic in nature; however, the composition of the interaction hyperpolarizabilities is much more complex. The overlap effects substantially quench the contributions due to electrostatic interactions, and therefore, the major components are due to the induction and exchange induction terms, as well as the intramolecular electron-correlation corrections. A particularly intriguing observation is that the interaction first hyperpolarizability in the studied systems not only is much larger than the corresponding sum of monomer properties, but also has the opposite sign. We show that this effect can be viewed as a direct consequence of hydrogen-bonding interactions that lead to a decrease of the hyperpolarizability of the proton acceptor and an increase of the hyperpolarizability of the proton donor. In the case of the first hyperpolarizability, we also observed the largest nonadditivity of interaction properties (nearly 17%) which further enhances the effects of pairwise interactions.

Fock space for fermion-like lattices and the linear Glauber model

The concept of Fock space representation is developed to deal with stochastic spin lattices written in terms of fermion operators. A density operator is introduced in order to follow in parallel the developments of the case of bosons in the literature. Some general conceptual quantities for spin lattices are then derived, including the notion of generating function and path integral via Grassmann variables. The formalism is used to derive the Liouvillian of the d-dimensional Linear Glauber dynamics in the Fock-space representation. Then the time evolution equations for the magnetization and the two-point correlation function are derived in terms of the number operator. (C) 2008 Elsevier B.V. All rights reserved.

Magnetic linear dichroism in absorption spectroscopy of EuTe

The magnetic linear dichroism (MLD) at band-edge photon energies in the Voigt geometry was calculated for EuTe. At the spin-flop transition, MLD shows a step-like increase. Above the spin-flop transition MLD slowly decreases and becomes zero when the averaged electronic charge becomes symmetric relative to the axis of light propagation. Further increase of the magnetic field causes ferromagnetic alignment of the spins along the magnetic field direction, and MLD is recovered but with an opposite sign, and reaches maximum absolute values. These results are explained by the rearrangement of the Eu(2+) spin distribution in the crystal lattice as a function of magnetic field, due to the Zeeman interaction, demonstrating that MLD can be a sensitive probe of the spin order in EuTe, and provides information that is not accessible from other magneto-optical techniques, such as magnetic circular dichroism measurement studies.

Linear covariant gauges on the lattice

Linear covariant gauges, such as Feynman gauge, are very useful in perturbative calculations. Their non-perturbative formulation is, however, highly non-trivial. In particular, it is a challenge to define linear covariant gauges on a lattice. We consider a class of gauges in lattice gauge theory that coincides with the perturbative definition of linear covariant gauges in the formal continuum limit. The corresponding gauge-fixing procedure is described and analyzed in detail, with an application to the pure SU(2) case. In addition, results for the gluon propagator in the two-dimensional case are given. (C) 2008 Elsevier B.V. All rights reserved.

The Itajai foreland basin: a tectono-sedimentary record of the Ediacaran period, Southern Brazil

The Itajai Basin located in the southern border of the Luis Alves Microplate is considered as a peripheral foreland basin related to the Dom Feliciano Belt. It presents an excellent record of the Ediacaran period, and its upper parts display the best Brazilian example of Precambrian turbiditic deposits. The basal succession of Itajai Group is represented by sandstones and conglomerates (BaA(0) Formation) deposited in alluvial and deltaic-fan systems. The marine upper sequences correspond to the Ribeiro Carvalho (channelized and non-channelized proximal silty-argillaceous rhythmic turbidites), Ribeiro Neisse (arkosic sandstones and siltites), and Ribeiro do Bode (distal silty turbidites) formations. The ApiA(0)na Formation felsic volcanic rocks crosscut the sedimentary succession. The Cambrian Subida leucosyenogranite represents the last felsic magmatic activity to affect the Itajai Basin. The Brusque Group and the Florianpolis Batholith are proposed as source areas for the sediments of the upper sequence. For the lower continental units the source areas are the Santa Catarina, So Miguel and CamboriA(0) complexes. The lack of any oceanic crust in the Itajai Basin suggests that the marine units were deposited in a restricted, internal sea. The sedimentation started around 600 Ma and ended before 560 Ma as indicated by the emplacement of rhyolitic domes. The Itajai Basin is temporally and tectonically correlated with the Camaqu Basin in Rio Grande do Sul and the Arroyo del Soldado/Piriapolis Basin in Uruguay. It also has several tectono-sedimentary characteristics in common with the African-equivalent Nama Basin.

30 million years of Permian volcanism recorded in the Choiyoi igneous province (W Argentina) and their source for younger ash fall deposits in the Parana Basin: SHRIMP U-Pb zircon geochronology evidence

We present four SHRIMP U-Pb zircon ages for the Choiyoi igneous province from the San Rafael Block, central-western Argentina. Dated samples come from the Yacimiento Los Reyunos Formation (281.4 +/- 2.5 Ma) of the Cochico Group (Lower Choiyoi section: andesitic breccias, dacitic to rhyolitic ignimbrites and continental conglomerates). Agua de los Burros Formation (264.8 +/- 2.3 Ma and 264.5 +/- 3.0 Ma) and Cerro Carrizalito Formation (251.9 +/- 2.7 Ma Upper Choiyoi section: rhyolitic ignimbrites and pyroclastic flows) spanning the entire Permian succession of the Choiyoi igneous province. A single ziron from the El Imperial Formation, that is overlain unconformably by the Choiyoi succession, yielded an early Permian age (297.2 +/- 5.3 Ma). while the main detrital zircon population indicated an Ordovician age (453.7 +/- 8.1 Ma). The new data establishes a more precise Permian age (Artinskian-Lopingian) for the section studied spanning 30 Ma of volcanic activity. Volcanological observations for the Choiyoi succession support the occurrence of explosive eruptions of plinian to ultraplinian magnitudes, capable of injecting enormous volumes of tephra in the troposphere-stratosphere. The new SHRIMP ages indicate contemporaneity between the Choyoi succession and the upper part of the Parana Basin late Paleozoic section, from the Irad up to the Rio do Rasto formations, encompassing about 24 Ma. Geochemical data show a general congruence in compositional and tectonic settings between the volcanics and Parana Basin Permian ash fall derived layers of bentonites. Thickness and granulometry of ash fall layers broadly fit into the depletion curve versus distance from the remote source vent of ultraplinian eruptions. Thus, we consider that the Choiyoi igneous province was the source of ash fall deposits in the upper Permian section of the Parana Basin. Data presented here allow a more consistent correlation between tectono-volcanic Permian events along the paleo-Pacific margin of southwestern Gondwana and the geological evolution of neighboring Paleozoic foreland basins in South America and Africa. (C) 2010 International Association for Gondwana Research. Published by Elsevier B.V. All rights reserved.

One, two or no record of late neoproterozoic glaciation in South-East Cameroon?

Severe climate changes culminating in at least three major glacial events have been recognized in the Neoproterozoic sedimentary record from many parts of the world Supportive to the global nature of these climatic shifts a considerable amount of data have been acquired from deposits exposed in Pan-African orogenic belts in southwestern and western Africa By comparison published data from the Pan-African belts in Central Africa are scarce We report here evidence of possibly two glacial events recorded in the Mintom Formation that is located on the margin of the Pan-African orogenic Yaounde belt in South-East Cameroon In the absence of reliable radiometric data only maximum and minimum age limits of 640 and 580 Ma respectively can at present be applied to the Mintom Formation The formation consists of two lithostratigraphic ensembles each subdivided in two members (i e in ascending stratigraphic order the Kol Metou Momibole and Atog Adjap Members) The basal ensemble exhibits a typical glacial to post-glacial succession It includes diamictites comprising cobbles and boulders in a massive argillaceous siltstone matrix and laminated siltstones followed by in sharp contact a 2 m-thick massive dolostone that yielded negative delta(13)C values (<-3 parts per thousand. V-PDB) similar to those reported for Marinoan cap carbonates elsewhere However uncertainty remains regarding the glacial influence on the siliciclastic facies because the diamictite is better explained as a mass-flow deposit and diagnostic features such as dropstones have not been seen in the overlying siltstones The Mintom Formation may thus provide an example of an unusual succession of non-glacial diamictite overlain by a truly glacial melt-related cap-carbonate We also report the recent discovery of ice-striated pavements on the structural surface cut in the Mintom Formation suggesting that glaciers developed after the latter had been deposited and deformed during the Pan-African orogeny Striations which consistently exhibit two principal orientations (N60 and N110) were identified in two different localities in the west of the study area on siltstones of the Kol Member and in the east on limestones of the Atog Adjap Member respectively N60-oriented striae indicate ice flow towards the WSW Assigning an age to these features remains problematical because they were not found associated with glaciogenic deposits Two hypotheses can equally be envisaged e either the striated surfaces are correlated (1) to the Gaskiers (or Neoproterozoic post-Gaskiers) glaciation and represent the youngest Ediacaran glacial event documented in the southern Yaounde belt or (2) to the Late Ordovician Hirnantian (Saharan) glaciation thereby providing new data about Hirnantian ice flows in Central Africa (C) 2010 Elsevier Ltd All rights reserved

Mineralogy and trace-element geochemistry of the high-grade iron ores of the Aguas Claras Mine and comparison with the Capao Xavier and Tamandua iron ore deposits, Quadrilahero Ferrifero, Brazil

Several major iron deposits occur in the Quadrilatero Ferrifero (QF), southeastern region of Brazil, where metamorphosed and heterogeneously deformed banded iron formation (BIF) of the Caue Formation, regionally called itabirite, was transformed into high- (Fe >64%) and lowgrade (30%deposits than siliceous itabirite. Field relations at the Aguas Claras and Capao Xavier deposits suggest that it is not possible to form huge soft high-grade supergene deposits from siliceous itabirite, unless another control, such as impermeable barriers, had played an important role. The occurrence in the Tamandua Mine of a large, soft, high-grade orebody formed from siliceous itabirite and closely associated with hypogene hard ore suggests that large, soft, high-grade orebodies of the Quadrilatero Ferrifero, which occur within siliceous itabirite, have a hypogene contribution in their formation.