This thesis presents a comparison for modeling and forecasting Chinese futures market of copper and aluminum with single time series and multivariate time series under linear restrictions. For single time series, data transformation for stationary purpose has been tested and performed before ARIMA model was built. For multivariate time series, co-integration rank test has been performed and included before VECM model was built. Based on selected models, the forecasting shows multivariate time series analysis has a better result than single time series, which indicates utilizing the relationships among the series can improve the accuracy of time series forecasting.

Thesis (Ph. D.)--Oregon State University, 2008. / Printout. Includes bibliographical references (leaves 124-133). Also available on the World Wide Web.

Univ., Habil.-Schr. u.d.T.: Gohm, Rolf: Elements of a spatial theory for noncommutative stationary processes with discrete time index--Greifswald, 2002. / Literaturverz. S. [165] - 168.

Thesis (Ph. D.)--Florida State University, 2003. / Advisor: Dr. John G. Dorsey, Florida State University, College of Arts and Sciences, Department of Chemistry and Biochemistry. Title and description from dissertation home page (viewed Aug. 24, 2004). Includes bibliographical references.

Systems of large sparse linear equations frequently arise in engineering and science. Therefore, there is a great need for methods that can solve these systems. In this thesis we will present three of the earliest and simplest iterative methods and also look at two more sophisticated methods. We will study their rate of convergence and illustrate them with examples.

Orientador: Isabel Cristina Sales Fontes Jardim / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Quimica / Made available in DSpace on 2018-08-06T15:47:25Z (GMT). No. of bitstreams: 1 Lourenco_Josimara_M.pdf: 719895 bytes, checksum: c0f8473e3a648286f7a03cc8d8056bc9 (MD5) Previous issue date: 2005 / Resumo: Em cromatografia líquida de alta eficiência (CLAE), a coluna é freqüentemente considerada como o "coração do sistema cromatográfico", pois é nela que ocorre a separação. Dessa forma, a busca por fases estacionárias capazes de realizar separações com alta eficiência e desempenho tem sido uma constante em muitos laboratórios de pesquisa. Polissiloxanos fenil substituídos têm sido extensivamente utilizados em cromatografia gasosa, mas a sua aplicação em CLAE tem despertado pouca atenção. Neste trabalho foram preparadas fases estacionárias para CLAE, com grupos fenil, a partir da sorção e subseqüente imobilização por tratamento térmico do copolímero de dimetil(52-48%)-metil-fenil(48-52%) siloxano sobre a sílica (Kromasil, 5 mm). Medidas da quantidade de carbono, volume de poro, área superficial específica, espectroscopia no infravermelho e RMN de C e Si foram realizadas para caracterizar físico-quimicamente as fases estacionárias. Também foram feitas avaliações cromatográficas utilizando misturas teste de compostos orgânicos de diferentes polaridades. As fases estacionárias preparadas a partir do copolímero de dimetil(52-48%)-metil-fenil(48-52%) siloxano sobre a sílica e imobilizadas termicamente a 150 °C, por 4,5 horas apresentaram alta eficiência (77000 N m), picos simétricos (As = 0,9) e bom desempenho cromatográfico. Os parâmetros cromatográficos indicaram que o tratamento térmico melhorou o desempenho das fases estacionárias e proporcionou um aumento da estabilidade frente a fases móveis neutras e alcalinas / Abstract: In high-performance liquid chromatography (HPLC), the column is frequently considered as the "heart of the chromatographic system", because is in the column that the separation occurs. In this way, the search for stationary phases able to achieve separations with high efficiency and performance has been a constant in many research laboratories. Phenyl substituted polysiloxanes have been used extensively in gas chromatography, but their application in HPLC has received little attention. In this work were prepared stationary phases for HPLC, with phenyl groups, using sorption and subsequent thermal immobilization of the copolymer of dimethyl(52-48%)-methyl-phenyl(48-52%) siloxane on silica (Kromasil, 5 mm). Measurement of amount of carbon, pore volume, specific surface area, infrared spectrum and NMR of C and Si were used to physical-chemically characterize the stationary phases. Chromatographic evaluations using test mixtures of organic compounds of different polarity were also carried out. The stationary phases prepared from the copolymer of dimethyl(52-48%)-methyl-phenyl(48-52%) siloxane on silica gel and thermally immobilized for 4.5 h at 150 °C show high efficience (77000 N m), symmetrical peaks (As = 0.9) and good chromatography performance. The chromatographic parameters show that the thermal treatment improved the performance of stationary phases and provided an increase of stability in neutral and basic mobile phases / Mestrado / Quimica Analitica / Mestre em Química

Doctor en Ciencias de la Ingeniería, Mención Modelación Matemática / Este trabajo de tesis contiene dos cap\'itulos principales, donde se estudian dos problemas independientes de Modelaci\'on Matem\'atica. En el Cap\'titulo 1 se estudia la existencia y unicidad de distribuciones quasi estacionarias para un movimiento Browniano con drift extinguido en cero, para el caso que infinito es Frontera de Entrada y cero frontera de salida de acuerdo a la clasificaci\'on de Feller. El trabajo est\'a relacionado con la publicaci\'on pionera \cite, donde algunas condiciones suficientes son establecidas para demostrar la existencia y unicidad de QSD en el contexto de una familia de Modelos de Din\'amica de Poblaciones y difusiones de Feller. El trabajo generaliza los teoremas m\'as importantes de \cite , ya que no se imponen condiciones extras para obtener los resultados de existencia y unicidad de QSD y la existencia del l\'imite de Yaglom. La parte t\'ecnica est\'a basada en la teor\'ia del problema de Sturm Liouville sobre la semirecta positiva. Espec\'ificamente, se demuestra que bajo las principales hip\'otesis existe espectro discreto si y solo si infinito es frontera de entrada y todas las eigenfunciones son simples e integrables respecto a la medida de rapidez del proceso.\\ En el cap\'itulo 2, se estudia el problema de obtener cotas optimales sobre el Hamiltoniano para el Modelo de Ising de largo alcance, con t\'ermino de interacci\'on decayendo de acuerdo a $d^{\alpha-2}$, $\alpha \in [0,1)$. El trabajo est\'a basado en el art\'iculo publicado en 2005 \cite{Paper1}, donde cotas optimales son obtenidas para el caso $\alpha \in [0,\frac{log{3}}{\log{2}}-1)$ en t\'erminos de estructuras jer\'arquicas llamadas tri\'angulos y contornos. Los teoremas principales de este trabajo pueden ser resumidos como (i) No existe una cota optimal para el Hamiltoniano en t\'erminos de tri\'angulos para $\alpha \in [\frac-1,1)$. (ii) Existe una cota optimal para el Hamiltoniano en t\'erminos de Contornos para $\alpha \in [0,1)$, resultados que son demostrado en los Teoremas \ref{XYZ} y \ref{Teo1} respectivamente. Ambos generalizan los resultados existentes, y constituyen la principal contribuci\'on de este trabajo. Para demostrar el Teorema \ref, se construye expl\'icitamente una familia de contraejemplos. La parte t\'ecnica est\'a fuertemente basada en la teor\'ia de Fractales sobre Conjuntos Discretos. Para demostrar teorema \ref{Teo1} , se usa el argumento se agrupar y sumar sobre contornos con la misma masa. Las demostraciones para ambos resultados son muy t\'ecnicas y requieren una gran cantidad de c\'alculos , los cuales son entregados en detalle. Por otra parte, los teoremas principales tienen importantes implicancias en \'esta clase de Modelos. La m\'as importante y directa es la existencia de una fase de transici\'on para bajas temperaturas basada en el argumento de Peierls. Dicha demostraci\'un, es tambi\'en entregada en este trabajo.

Multivariate time series analysis became popular in the early 1950s when the need to analyse time series simultaneously arose in the field of economics. This study provides an overview of some of the aspects of multivariate time series analysis in the case of stationarity. The VARMA (vector autoregressive moving average) class of multivariate time series models, including pure vector autoregressive (VAR) and vector moving average (VMA) models is considered. Methods based on moments and information criteria for the determination of the appropriate order of a model suitable for an observed multivariate time series are discussed. Feasible methods of estimation based on the least squares and/or maximum likelihood are provided for the different types of VARMA models. In some cases, the estimation is more complicated due to the identification problem and the nonlinearity of the normal equations. It is shown that the significance of individual estimates can be established by using hypothesis tests based on the asymptotic properties of the estimators. Diagnostic tests for the adequacy of the fitted model are discussed and illustrated. These include methods based on both univariate and multivariate procedures. The complete model building process is illustrated by means of case studies on multivariate electricity demand and temperature time series. Throughout the study numerical examples are used to illustrate concepts. Computer program code (using basic built-in multivariate functions) is given for all the examples. The results are benchmarked against those produced by a dedicated procedure for multivariate time series. It is envisaged that the program code (given in SAS/IML) could be made available to a much wider user community, without much difficulty, by translation into open source platforms. / Dissertation (MSc (Mathematical Statistics))--University of Pretoria, 2008. / Mathematics and Applied Mathematics / unrestricted

Extreme winter weather events in North America have become more frequent and increasingly destructive. This phenomenon was linked to a jet stream pattern that generates abnormally warm conditions in the west and cold conditions in the east, referred to as the North American Winter Dipole. Studies have shown that the Dipole may have amplified and this amplification could be linked to global warming. By analyzing the atmospheric and oceanic data worldwide, the wintertime circulation in the Northern Hemisphere shows signs of a persistent change after the 1980s. In the first part of this study, we examine how the ocean has changed in correspondence to the Dipole and the evolution of the pattern change. In the second part of this study, we use multiple global reanalysis datasets to construct the Dipole index. The result validates the reported Dipole variation during the modern period. We also use the Dipole variance to investigate the Dipole’s behavior in the paleoclimate and future warming conditions. Overall, we sought to better understand how the Dipole pattern evolves and how it may link to the different forcing, as a way to anticipate future change in North America’s winter.

Inventory control problems constitute one of the most important research problems due to their connection with real life applications. Naturally, real life is full of uncertainty so are the most of the inventory problems. Unfortunately, it is a very challenging task to manage inventories effectively especially under uncertainty. This dissertation mainly deals with single-item, periodic review, and stochastic dynamic inventory control problems particularly on replenishment cycle control rule known as the (R, S) policy. Contribution of this thesis is multiold. In each chapter a particular research question is investigated. At the end of the day, we will be showing that non-stationary (R, S) policies are indispensable not only for its cost efficiency but its effectiveness and practicality. More specifically, the non-stationary (R, S) policy provides a convenient, efficient, effective, and modular solution for non-stationary stochastic inventory control problems.