A nonparametric ensemble transform method for Bayesian inference

Many applications, such as intermittent data assimilation, lead to a recursive application of Bayesian inference within a Monte Carlo context. Popular data assimilation algorithms include sequential Monte Carlo methods and ensemble Kalman filters (EnKFs). These methods differ in the way Bayesian inference is implemented. Sequential Monte Carlo methods rely on importance sampling combined with a resampling step, while EnKFs utilize a linear transformation of Monte Carlo samples based on the classic Kalman filter. While EnKFs have proven to be quite robust even for small ensemble sizes, they are not consistent since their derivation relies on a linear regression ansatz. In this paper, we propose another transform method, which does not rely on any a priori assumptions on the underlying prior and posterior distributions. The new method is based on solving an optimal transportation problem for discrete random variables. © 2013, Society for Industrial and Applied Mathematics

Measurement of the top quark mass in the lepton plus jets final state with the matrix element method

We present a measurement of the top quark mass with the matrix element method in the lepton+jets final state. As the energy scale for calorimeter jets represents the dominant source of systematic uncertainty, the matrix element likelihood is extended by an additional parameter, which is defined as a global multiplicative factor applied to the standard energy scale. The top quark mass is obtained from a fit that yields the combined statistical and systematic jet energy scale uncertainty. Using a data set of 0.4 fb(-1) taken with the D0 experiment at Run II of the Fermilab Tevatron Collider, the mass of the top quark is measured using topological information to be: m(top)(center dot+jets)(topo)=169.2(-7.4)(+5.0)(stat+JES)(-1.4)(+1.5)(syst) GeV, and when information about identified b jets is included: m(top)(center dot+jets)(b-tag)=170.3(-4.5)(+4.1)(stat+ JES)(-1.8)(+1.2)(syst) GeV. The measurements yield a jet energy scale consistent with the reference scale.

Integral transform method for laminar heat transfer convection of herschel-bulkley fluids within concentric annular ducts

ABSTRACT: Related momentum and energy equations describing the heat and fluid flow of Herschel-Bulkley fluids within concentric annular ducts are analytically solved using the classical integral transform technique, which permits accurate determination of parameters of practical interest in engineering such as friction factors and Nusselt numbers for the duct length. In analyzing the problem, thermally developing flow is assumed and the duct walls are subjected to boundary conditions of first kind. Results are computed for the velocity and temperature fields as well as for the parameters cited above with different power-law indices, yield numbers and aspect ratios. Comparisons are also made with previous work available in the literature, providing direct validation of the results and showing that they are consistent.

Improvements in the score matrix calculation method using parallel score estimating algorithm

The increasing amount of sequences stored in genomic databases has become unfeasible to the sequential analysis. Then, the parallel computing brought its power to the Bioinformatics through parallel algorithms to align and analyze the sequences, providing improvements mainly in the running time of these algorithms. In many situations, the parallel strategy contributes to reducing the computational complexity of the big problems. This work shows some results obtained by an implementation of a parallel score estimating technique for the score matrix calculation stage, which is the first stage of a progressive multiple sequence alignment. The performance and quality of the parallel score estimating are compared with the results of a dynamic programming approach also implemented in parallel. This comparison shows a significant reduction of running time. Moreover, the quality of the final alignment, using the new strategy, is analyzed and compared with the quality of the approach with dynamic programming.

Search for a standard model Higgs boson produced in association with a top-quark pair and decaying to bottom quarks using a matrix element method

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Analysis Methodology for Vortex-Induced Motion (VIM) of a Monocolumn Platform Applying the Hilbert-Huang Transform Method

Vortex-induced motion (VIM) is a highly nonlinear dynamic phenomenon. Usual spectral analysis methods, using the Fourier transform, rely on the hypotheses of linear and stationary dynamics. A method to treat nonstationary signals that emerge from nonlinear systems is denoted Hilbert-Huang transform (HHT) method. The development of an analysis methodology to study the VIM of a monocolumn production, storage, and offloading system using HHT is presented. The purposes of the present methodology are to improve the statistics analysis of VIM. The results showed to be comparable to results obtained from a traditional analysis (mean of the 10% highest peaks) particularly for the motions in the transverse direction, although the difference between the results from the traditional analysis for the motions in the in-line direction showed a difference of around 25%. The results from the HHT analysis are more reliable than the traditional ones, owing to the larger number of points to calculate the statistics characteristics. These results may be used to design risers and mooring lines, as well as to obtain VIM parameters to calibrate numerical predictions. [DOI: 10.1115/1.4003493]

Study of electromagnetic reactions on light nuclei with the Lorentz Integral Transform method

In dieser Arbeit aus dem Bereich der Wenig-Nukleonen-Physik wird die neu entwickelte Methode der Lorentz Integral Transformation (LIT) auf die Untersuchung von Kernphotoabsorption und Elektronenstreuung an leichten Kernen angewendet. Die LIT-Methode ermoeglicht exakte Rechnungen durchzufuehren, ohne explizite Bestimmung der Endzustaende im Kontinuum. Das Problem wird auf die Loesung einer bindungzustandsaehnlichen Gleichung reduziert, bei der die Endzustandswechselwirkung vollstaendig beruecksichtigt wird. Die Loesung der LIT-Gleichung wird mit Hilfe einer Entwicklung nach hypersphaerischen harmonischen Funktionen durchgefuehrt, deren Konvergenz durch Anwendung einer effektiven Wechselwirkung im Rahmem des hypersphaerischen Formalismus (EIHH) beschleunigt wird. In dieser Arbeit wird die erste mikroskopische Berechnung des totalen Wirkungsquerschnittes fuer Photoabsorption unterhalb der Pionproduktionsschwelle an 6Li, 6He und 7Li vorgestellt. Die Rechnungen werden mit zentralen semirealistischen NN-Wechselwirkungen durchgefuehrt, die die Tensor Kraft teilweise simulieren, da die Bindungsenergien von Deuteron und von Drei-Teilchen-Kernen richtig reproduziert werden. Der Wirkungsquerschnitt fur Photoabsorption an 6Li zeigt nur eine Dipol-Riesenresonanz, waehrend 6He zwei unterschiedliche Piks aufweist, die dem Aufbruch vom Halo und vom Alpha-Core entsprechen. Der Vergleich mit experimentellen Daten zeigt, dass die Addition einer P-Wellen-Wechselwirkung die Uebereinstimmung wesentlich verbessert. Bei 7Li wird nur eine Dipol-Riesenresonanz gefunden, die gut mit den verfuegbaren experimentellen Daten uebereinstimmt. Bezueglich der Elektronenstreuung wird die Berechnung der longitudinalen und transversalen Antwortfunktionen von 4He im quasi-elastischen Bereich fuer mittlere Werte des Impulsuebertrages dargestellt. Fuer die Ladungs- und Stromoperatoren wird ein nichtrelativistisches Modell verwendet. Die Rechnungen sind mit semirealistischen Wechselwirkungen durchgefuert und ein eichinvarianter Strom wird durch die Einfuehrung eines Mesonaustauschstroms gewonnen. Die Wirkung des Zweiteilchenstroms auf die transversalen Antwortfunktionen wird untersucht. Vorlaeufige Ergebnisse werden gezeigt und mit den verfuegbaren experimentellen Daten verglichen.

A solvent-free matrix application method for matrix-assisted laser desorption/ionization imaging of small molecules

Matrix application continues to be a critical step in sample preparation for matrix-assisted laser desorption/ionization (MALDI) mass spectrometry imaging (MSI). Imaging of small molecules such as drugs and metabolites is particularly problematic because the commonly used washing steps to remove salts are usually omitted as they may also remove the analyte, and analyte spreading is more likely with conventional wet matrix application methods. We have developed a method which uses the application of matrix as a dry, finely divided powder, here referred to as dry matrix application, for the imaging of drug compounds. This appears to offer a complementary method to wet matrix application for the MALDI-MSI of small molecules, with the alternative matrix application techniques producing different ion profiles, and allows the visualization of compounds not observed using wet matrix application methods. We demonstrate its value in imaging clozapine from rat kidney and 4-bromophenyl-1,4-diazabicyclo(3.2.2)nonane-4-carboxylic acid from rat brain. In addition, exposure of the dry matrix coated sample to a saturated moist atmosphere appears to enhance the visualization of a different set of molecules.

Advanced fibre gratings and their applications

This thesis describes a detailed study of advanced fibre grating devices using Bragg (FBG) and long-period (LPG) structures and their applications in optical communications and sensing. The major contributions presented in this thesis are summarised below. One of the most important contributions from the research work presented in this thesis is a systematic theoretical study of many distinguishing structures of fibre gratings. Starting from the Maxwell equations, the coupled-mode equations for both FBG and LPG were derived and the mode-overlap factor was analytically discussed. Computing simulation programmes utilising matrix transform method based on the models built upon the coupled-mode equations were developed, enabling simulations of spectral response in terms of reflectivity, bandwidth, sidelobes and dispersion of gratings of different structures including uniform and chirped, phase-shifted, Moiré, sampled Bragg gratings, phase-shifted and cascaded long-period gratings. Although the majority of these structures were modelled numerically, analytical expressions for some complex structures were developed with a clear physical picture. Several apodisation functions were proposed to improve sidelobe suppression, which guided effective production of practical devices for demanding applications. Fibre grating fabrication is the other major part involved in the Ph.D. programme. Both the holographic and scan-phase-mask methods were employed to fabricate Bragg and long-period gratings of standard and novel structures. Significant improvements were particularly made in the scan-phase-mask method to enable the arbitrarily tailoring of the spectral response of grating devices. Two specific techniques - slow-shifting and fast-dithering the phase-mask implemented by a computer controlled piezo - were developed to write high quality phase-shifted, sampled and apodised gratings. A large number of LabVIEW programmes were constructed to implement standard and novel fabrication techniques. In addition, some fundamental studies of grating growth in relating to the UV exposure and hydrogenation induced index were carried out. In particular, Type IIa gratings in non-hydrogenated B/Ge co-doped fibres and a re-generated grating in hydrogenated B/Ge fibre were investigated, showing a significant observation of thermal coefficient reduction. Optical sensing applications utilising fibre grating devices form the third major part of the research work presented in this thesis. Several experiments of novel sensing and sensing-demodulating were implemented. For the first time, an intensity and wavelength dual-coding interrogation technique was demonstrated showing significantly enhanced capacity of grating sensor multiplexing. Based on the mode-splitting measurement, instead of using conventional wavelength-shifting detection technique, successful demonstrations were also made for optical load and bend sensing of ultra-high sensitivity employing LPG structures. In addition, edge-filters and low-loss high-rejection bandpass filters of 50nm stop-band were fabricated for application in optical sensing and high-speed telecommunication systems

On the relationship between a 2 x 2 matrix and second-order scalar spectral problems for integrable equations

A simple proof is given that a 2 x 2 matrix scheme for an inverse scattering transform method for integrable equations can be converted into the standard form of the second-order scalar spectral problem associated with the same equations. Simple formulae relating these two kinds of representation of integrable equations are established.

Development of the optimal estimation method for inversion of stratospheric emission spectra and validation of HNO3 profiles

This doctoral thesis focuses on ground-based measurements of stratospheric nitric acid (HNO3)concentrations obtained by means of the Ground-Based Millimeter-wave Spectrometer (GBMS). Pressure broadened HNO3 emission spectra are analyzed using a new inversion algorithm developed as part of this thesis work and the retrieved vertical profiles are extensively compared to satellite-based data. This comparison effort I carried out has a key role in establishing a long-term (1991-2010), global data record of stratospheric HNO3, with an expected impact on studies concerning ozone decline and recovery. The first part of this work is focused on the development of an ad hoc version of the Optimal Estimation Method (Rodgers, 2000) in order to retrieve HNO3 spectra observed by means of GBMS. I also performed a comparison between HNO3 vertical profiles retrieved with the OEM and those obtained with the old iterative Matrix Inversion method. Results show no significant differences in retrieved profiles and error estimates, with the OEM providing however additional information needed to better characterize the retrievals. A final section of this first part of the work is dedicated to a brief review on the application of the OEM to other trace gases observed by GBMS, namely O3 and N2O. The second part of this study deals with the validation of HNO3 profiles obtained with the new inversion method. The first step has been the validation of GBMS measurements of tropospheric opacity, which is a necessary tool in the calibration of any GBMS spectra. This was achieved by means of comparisons among correlative measurements of water vapor column content (or Precipitable Water Vapor, PWV) since, in the spectral region observed by GBMS, the tropospheric opacity is almost entirely due to water vapor absorption. In particular, I compared GBMS PWV measurements collected during the primary field campaign of the ECOWAR project (Bhawar et al., 2008) with simultaneous PWV observations obtained with Vaisala RS92k radiosondes, a Raman lidar, and an IR Fourier transform spectrometer. I found that GBMS PWV measurements are in good agreement with the other three data sets exhibiting a mean difference between observations of ~9%. After this initial validation, GBMS HNO3 retrievals have been compared to two sets of satellite data produced by the two NASA/JPL Microwave Limb Sounder (MLS) experiments (aboard the Upper Atmosphere Research Satellite (UARS) from 1991 to 1999, and on the Earth Observing System (EOS) Aura mission from 2004 to date). This part of my thesis is inserted in GOZCARDS (Global Ozone Chemistry and Related Trace gas Data Records for the Stratosphere), a multi-year project, aimed at developing a long-term data record of stratospheric constituents relevant to the issues of ozone decline and expected recovery. This data record will be based mainly on satellite-derived measurements but ground-based observations will be pivotal for assessing offsets between satellite data sets. Since the GBMS has been operated for more than 15 years, its nitric acid data record offers a unique opportunity for cross-calibrating HNO3 measurements from the two MLS experiments. I compare GBMS HNO3 measurements obtained from the Italian Alpine station of Testa Grigia (45.9° N, 7.7° E, elev. 3500 m), during the period February 2004 - March 2007, and from Thule Air Base, Greenland (76.5°N 68.8°W), during polar winter 2008/09, and Aura MLS observations. A similar intercomparison is made between UARS MLS HNO3 measurements with those carried out from the GBMS at South Pole, Antarctica (90°S), during the most part of 1993 and 1995. I assess systematic differences between GBMS and both UARS and Aura HNO3 data sets at seven potential temperature levels. Results show that, except for measurements carried out at Thule, ground based and satellite data sets are consistent within the errors, at all potential temperature levels.

Power Secant Method applied to natural frequency extraction of Timoshenko beam structures

This work deals with an improved plane frame formulation whose exact dynamic stiffness matrix (DSM) presents, uniquely, null determinant for the natural frequencies. In comparison with the classical DSM, the formulation herein presented has some major advantages: local mode shapes are preserved in the formulation so that, for any positive frequency, the DSM will never be ill-conditioned; in the absence of poles, it is possible to employ the secant method in order to have a more computationally efficient eigenvalue extraction procedure. Applying the procedure to the more general case of Timoshenko beams, we introduce a new technique, named ""power deflation"", that makes the secant method suitable for the transcendental nonlinear eigenvalue problems based on the improved DSM. In order to avoid overflow occurrences that can hinder the secant method iterations, limiting frequencies are formulated, with scaling also applied to the eigenvalue problem. Comparisons with results available in the literature demonstrate the strength of the proposed method. Computational efficiency is compared with solutions obtained both by FEM and by the Wittrick-Williams algorithm.

Predictive Optimal Matrix Converter Control for a Dynamic Voltage Restorer with Flywheel Energy Storage

This paper presents a predictive optimal matrix converter controller for a flywheel energy storage system used as Dynamic Voltage Restorer (DVR). The flywheel energy storage device is based on a steel seamless tube mounted as a vertical axis flywheel to store kinetic energy. The motor/generator is a Permanent Magnet Synchronous Machine driven by the AC-AC Matrix Converter. The matrix control method uses a discrete-time model of the converter system to predict the expected values of the input and output currents for all the 27 possible vectors generated by the matrix converter. An optimal controller minimizes control errors using a weighted cost functional. The flywheel and control process was tested as a DVR to mitigate voltage sags and swells. Simulation results show that the DVR is able to compensate the critical load voltage without delays, voltage undershoots or overshoots, overcoming the input/output coupling of matrix converters.

Improved estimation of the covariance matrix of stock returns with an application to portofolio selection

This paper proposes to estimate the covariance matrix of stock returnsby an optimally weighted average of two existing estimators: the samplecovariance matrix and single-index covariance matrix. This method isgenerally known as shrinkage, and it is standard in decision theory andin empirical Bayesian statistics. Our shrinkage estimator can be seenas a way to account for extra-market covariance without having to specifyan arbitrary multi-factor structure. For NYSE and AMEX stock returns from1972 to 1995, it can be used to select portfolios with significantly lowerout-of-sample variance than a set of existing estimators, includingmulti-factor models.