State detection in a financial portfolio: a self-organizing maps approach for financial time series

This study analyses financial data using the result characterization of a self-organized neural network model. The goal was prototyping a tool that may help an economist or a market analyst to analyse stock market series. To reach this goal, the tool shows economic dependencies and statistics measures over stock market series. The neural network SOM (self-organizing maps) model was used to ex-tract behavioural patterns of the data analysed. Based on this model, it was de-veloped an application to analyse financial data. This application uses a portfo-lio of correlated markets or inverse-correlated markets as input. After the anal-ysis with SOM, the result is represented by micro clusters that are organized by its behaviour tendency. During the study appeared the need of a better analysis for SOM algo-rithm results. This problem was solved with a cluster solution technique, which groups the micro clusters from SOM U-Matrix analyses. The study showed that the correlation and inverse-correlation markets projects multiple clusters of data. These clusters represent multiple trend states that may be useful for technical professionals.

Premature Osteoblast Clustering by Enamel Matrix Proteins Induces Osteoblast Differentiation through Up-Regulation of Connexin 43 and N-Cadherin

In recent years, enamel matrix derivative (EMD) has garnered much interest in the dental field for its apparent bioactivity that stimulates regeneration of periodontal tissues including periodontal ligament, cementum and alveolar bone. Despite its widespread use, the underlying cellular mechanisms remain unclear and an understanding of its biological interactions could identify new strategies for tissue engineering. Previous in vitro research has demonstrated that EMD promotes premature osteoblast clustering at early time points. The aim of the present study was to evaluate the influence of cell clustering on vital osteoblast cell-cell communication and adhesion molecules, connexin 43 (cx43) and N-cadherin (N-cad) as assessed by immunofluorescence imaging, real-time PCR and Western blot analysis. In addition, differentiation markers of osteoblasts were quantified using alkaline phosphatase, osteocalcin and von Kossa staining. EMD significantly increased the expression of connexin 43 and N-cadherin at early time points ranging from 2 to 5 days. Protein expression was localized to cell membranes when compared to control groups. Alkaline phosphatase activity was also significantly increased on EMD-coated samples at 3, 5 and 7 days post seeding. Interestingly, higher activity was localized to cell cluster regions. There was a 3 fold increase in osteocalcin and bone sialoprotein mRNA levels for osteoblasts cultured on EMD-coated culture dishes. Moreover, EMD significantly increased extracellular mineral deposition in cell clusters as assessed through von Kossa staining at 5, 7, 10 and 14 days post seeding. We conclude that EMD up-regulates the expression of vital osteoblast cell-cell communication and adhesion molecules, which enhances the differentiation and mineralization activity of osteoblasts. These findings provide further support for the clinical evidence that EMD increases the speed and quality of new bone formation in vivo.

Liberia : need to improve accountability and control over U.S. assistance : report to the Edward M. Kennedy, U.S. Senate /

"B-226281."

Sparse matrix multiplication on a reconfigurable many-core architecture

Sparse matrix-vector multiplication (SMVM) is a fundamental operation in many scientific and engineering applications. In many cases sparse matrices have thousands of rows and columns where most of the entries are zero, while non-zero data is spread over the matrix. This sparsity of data locality reduces the effectiveness of data cache in general-purpose processors quite reducing their performance efficiency when compared to what is achieved with dense matrix multiplication. In this paper, we propose a parallel processing solution for SMVM in a many-core architecture. The architecture is tested with known benchmarks using a ZYNQ-7020 FPGA. The architecture is scalable in the number of core elements and limited only by the available memory bandwidth. It achieves performance efficiencies up to almost 70% and better performances than previous FPGA designs.

Unsupervised classification of remote sensing images combining Self Organizing Maps and segmentation techniques

Dissertation submitted in partial fulfillment of the requirements for the Degree of Master of Science in Geospatial Technologies.

Os efeitos das variações das taxas de juro na volatilidade dos mercados bolsistas

Mestrado em Controlo de Gestão e dos Negócios

Elliptical orbital studies of H21 and H2 molecules /|nS. K. Gupta. -- 260 St. Catharines, Ont. : [s. n.],

Calculations are performed on the \S <:Jd ground states of d ' + the H and HC) molecules using a basis set of non-integral ~ ~ I elliptical orbitals. Different variational wavefunctions constructed i- for H~ involved one parameter to three par~~eter variation. In order to l"'educe the ntunber of parameters in most commonly 0- used basis orbitals set, the importance of the term (,+~) Y\ over the term ;u 'Where n is a variational pararneter and the value of cr may be given by boundary condition or cusp condition is outlined in Chapters II and III. It is found that the two parameter -+ over ;U for the two electron problem, the variational energy is computed for the H~ molecule from unrestricted two parameter closed shell wavefunctions including the term U+ft)

High-strength, healable, supramolecular polymer nanocomposites

A supramolecular polymer blend, formed via π-π interactions between a π-electron rich pyrenyl endcapped oligomer and a chain-folding oligomer containing pairs of π-electron poor naphthalene-diimide (NDI) units, has been reinforced with cellulose nanocrystals (CNCs) to afford a healable nanocomposite material. Nanocomposites with varying weight percentage of CNCs (from 1.25 to 20.0 wt.%) within the healable supramolecular polymeric matrix have been prepared via solvent casting followed by compression molding, and their mechanical properties and healing behavior have been evaluated. It is found that homogeneously dispersed films can be formed with CNCs at less than 10 wt.%. Above 10 wt.% CNC heterogeneous nanocomposites were obtained. All the nanocomposites formed could be re-healed upon exposure to elevated temperatures although, for the homogeneous films, it was found that the healing rate was reduced with increasing CNC content. The best combination of healing efficiency and mechanical properties was obtained with the 7.5 wt.% CNC nanocomposite which exhibited a tensile modulus enhanced by as much as a factor of 20 over the matrix material alone and could be fully re-healed at 85 °C within 30 minutes. Thus it is demonstrated that supramolecular nanocomposites can afford greatly enhanced mechanical properties relative to the unreinforced polymer, while still allowing efficient thermal healing.

Controle preditivo robusto baseado em desigualdades matriciais lineares aplicado a um sistema de tanques acoplados

This work deals with an on-line control strategy based on Robust Model Predictive Control (RMPC) technique applied in a real coupled tanks system. This process consists of two coupled tanks and a pump to feed the liquid to the system. The control objective (regulator problem) is to keep the tanks levels in the considered operation point even in the presence of disturbance. The RMPC is a technique that allows explicit incorporation of the plant uncertainty in the problem formulation. The goal is to design, at each time step, a state-feedback control law that minimizes a 'worst-case' infinite horizon objective function, subject to constraint in the control. The existence of a feedback control law satisfying the input constraints is reduced to a convex optimization over linear matrix inequalities (LMIs) problem. It is shown in this work that for the plant uncertainty described by the polytope, the feasible receding horizon state feedback control design is robustly stabilizing. The software implementation of the RMPC is made using Scilab, and its communication with Coupled Tanks Systems is done through the OLE for Process Control (OPC) industrial protocol

High-resolution paleomagnetic investigation in the Iceland Basin

Based on discrete samples, we report new high-resolution records of the ~185 kyr Iceland Basin (IB) geomagnetic excursion from Ocean Drilling Project (ODP) Site 1063 on the Bermuda Rise (sedimentation rate 32 cm/kyr) and from ODP Site 983 in the far North Atlantic (sedimentation rate 18 cm/kyr). Two records from Holes 1063A and 1063B are very consistent, and provide the highest resolution of the detailed field behaviour during the IB excursion obtained so far. Inclination records from Holes 983B and 983C in the far North Atlantic are also very consistent, whereas declination anomalies deviate more notably. The pseudo-Thellier (PT) technique was applied along with more conventional palaeointensity proxies (NRM/ARM and NRM/kappa) to recover relative palaeointensity (RPI) estimates from Hole 1063A and Hole 983B. As expected, these proxies indicate that the field intensity generally dropped at both sites during the IB excursion, but also that the history of RPI from the two sites is different. VGPs from Site 1063 indicate that the field at this location experienced some stop-and-go behaviour between patches of intense vertical flux over North America and the tip of South America, areas which coincide fairly well with patches of preferred transitional VGP clustering from reversals and zones of high seismic velocity in the lower mantle. Changes in RPI at this location were generally gradual, possibly due to the proximity of these flux patches, and the first period of VGP-clustering over North America was accompanied by a conspicuous increase in RPI. VGPs from Site 983 track along a different path, and the associated RPI changes are very abrupt and completely synchronous with the onset and termination of the excursion. The differing VGP paths from Sites 1063 and 983 indicate that the global field structure during the IB excursion was not dominated by a single dipole.

Geochemistry of basement carbonates from ODP Site 801 in the western Pacific Ocean

Many studies argue, based partly on Pb isotopic evidence, that recycled, subducted slabs reside in the mantle source of ocean island basalts (OIB) (Hofmann and White, 1982, doi:10.1016/0012-821X(82)90161-3; Weaver, 1991 doi:10.1016/0012-821X(91)90217-6; Lassiter, and Hauri, 1998, doi:10.1016/S0012-821X(98)00240-4). Such models, however, have remained largely untested against actual subduction zone inputs, due to the scarcity of comprehensive measurements of both radioactive parents (Th and U) and radiogenic daughter (Pb) in altered oceanic crust (AOC). Here, we discuss new, comprehensive measurements of U, Th, and Pb concentrations in the oldest AOC, ODP Site 801, and consider the effect of subducting this crust on the long-term Pb isotope evolution of the mantle. The upper 500 m of AOC at Site 801 shows >4-fold enrichment in U over pristine glass during seafloor alteration, but no net change to Pb or Th. Without subduction zone processing, ancient AOC would evolve to low 208Pb/206Pb compositions unobserved in the modern mantle (Hart and Staudigel, 1989 [Isotopic characterization and identification of recycled components, in: Crust/Mantle Recycling at Convergence Zones, Eds. S.R. Hart, L. Gqlen, NATO ASI Series. Series C: Mathematical and Physical Sciences 258, pp. 15-28, D. Reidel Publishing Company, Dordrecht-Boston, 1989]). Subduction, however, drives U-Th-Pb fractionation as AOC dehydrates in the earth's interior. Pacific arcs define mixing trends requiring 8-fold enrichment in Pb over U in AOC-derived fluid. A mass balance across the Mariana subduction zone shows that 44-75% of Pb but <10% of U is lost from AOC to the arc, and a further 10-23% of Pb and 19-40% of U is lost to the back-arc. Pb is lost shallow and U deep from subducted AOC, which may be a consequence of the stability of phases binding these elements during seafloor alteration: U in carbonate and Pb in sulfides. The upper end of these recycling estimates, which reflect maximum arc and back-arc growth rates, remove enough Pb and U from the slab to enable it to evolve rapidly (<<0.5 Ga) to sources suitable to explain the 208Pb/206Pb isotopic array of OIB, although these conditions fail to simultaneously satisfy the 207Pb/206Pb system. Lower growth rates would require additional U loss (29%) at depths beyond the zones of arc and back-arc magmagenesis, which would decrease upper mantle kappa (232Th/238U) over time, consistent with one solution to the "kappa conundrum" (Elliott et al., 1999, doi:10.1016/S0012-821X(99)00077-1). The net effects of alteration (doubling of l [238U/204Pb]) and subduction (doubling of omega [232Th/204Pb]) are sufficient to create the Pb isotopic signatures of oceanic basalts.

Chick Harley, Ohio State halfback vs UM, 1919, UM Football

[Ohio All-American halfback Chic Harley scored on a 40-yard run in OSU's first victory in over U-M 1919. Michiganensian caption reads: "This is the play that gave Ohio the victory. Harley is making up his mind whether to pass to Stinchcomb or to plunge through Michigan's bewildered line."]

Computer simulation of radio-frequency methane/hydrogen plasmas and their interaction with GaAs Surfaces

The following thesis describes the computer modelling of radio frequency capacitively coupled methane/hydrogen plasmas and the consequences for the reactive ion etching of (100) GaAs surfaces. In addition a range of etching experiments was undertaken over a matrix of pressure, power and methane concentration. The resulting surfaces were investigated using X-ray photoelectron spectroscopy and the results were discussed in terms of physical and chemical models of particle/surface interactions in addition to the predictions for energies, angles and relative fluxes to the substrate of the various plasma species. The model consisted of a Monte Carlo code which followed electrons and ions through the plasma and sheath potentials whilst taking account of collisions with background neutral gas molecules. The ionisation profile output from the electron module was used as input for the ionic module. Momentum scattering interactions of ions with gas molecules were investigated via different models and compared against results given by quantum mechanical code. The interactions were treated as central potential scattering events and the resulting neutral cascades were followed. The resulting predictions for ion energies at the cathode compared well to experimental ion energy distributions and this verified the particular form of the electrical potentials used and their applicability in the particular geometry plasma cell used in the etching experiments. The final code was used to investigate the effect of external plasma parameters on the mass distribution, energy and angles of all species impingent on the electrodes. Comparisons of electron energies in the plasma also agreed favourably with measurements made using a Langmuir electric probe. The surface analysis showed the surfaces all to be depleted in arsenic due to its preferential removal and the resultant Ga:As ratio in the surface was found to be directly linked to the etch rate. The etch rate was determined by the methane flux which was predicted by the code.

Matrix-product codes over Fq

Codes C-1,...,C-M of length it over F-q and an M x N matrix A over F-q define a matrix-product code C = [C-1 (...) C-M] (.) A consisting of all matrix products [c(1) (...) c(M)] (.) A. This generalizes the (u/u + v)-, (u + v + w/2u + v/u)-, (a + x/b + x/a + b + x)-, (u + v/u - v)- etc. constructions. We study matrix-product codes using Linear Algebra. This provides a basis for a unified analysis of /C/, d(C), the minimum Hamming distance of C, and C-perpendicular to. It also reveals an interesting connection with MDS codes. We determine /C/ when A is non-singular. To underbound d(C), we need A to be 'non-singular by columns (NSC)'. We investigate NSC matrices. We show that Generalized Reed-Muller codes are iterative NSC matrix-product codes, generalizing the construction of Reed-Muller codes, as are the ternary 'Main Sequence codes'. We obtain a simpler proof of the minimum Hamming distance of such families of codes. If A is square and NSC, C-perpendicular to can be described using C-1(perpendicular to),...,C-M(perpendicular to) and a transformation of A. This yields d(C-perpendicular to). Finally we show that an NSC matrix-product code is a generalized concatenated code.