En esta tesis se estudia el Problema de Dise��o de Redes Continuo (CNDP por sus siglas en ingl��s) y se modela como un problema de programaci��n matem��tica binivel. El nivel superior se define como la suma total de los tiempos de viaje y costos de inversi��n de los incrementos de capacidad de las conexiones de la red, mientras que el nivel inferior se enfoca en el equilibrio de flujo de los usuarios que se modela como un problema de minimizaci��n. Se utilizan tres t��cnicas que no usan derivadas para obtener soluciones factibles del problema: 1) el m��todo de Nelder-Mead adaptado al CNDP, 2) un Metaheur��stico de B��squeda dispersa (BD), 3) el uso del algoritmo de Nelder-Mead en BD como un m��todo de mejora de soluciones. Los m��todos propuestos obtienen resultados que igualan a los que se encuentran en la literatura, aunque el m��todo de BD con Nelder-Mead mejora resultados previamente reportados para dos instancias de prueba del problema. / Palabras claves: CNDP, Programaci��n Binivel, Nelder-Mead, B��squeda Dispersa.

This thesis studies the problem of extending the concept of γ-active constraints to Convex Semi-Inαnite Programming. To achieve this goal, extensive knowledge of topology, convex analysis, real analysis and optimization is needed. We base ourselves on the deαnition and results shown in previous publications and present two approaches to extend this deαnition to the case of Convex Semi-Inαnite Programming. We also provide a comparison of the two approaches, where we state their limitations and advantages. / Key Words: Semi-Inαnite Programming, Extended Active Constraints, Convex Program-ming, Optimality Conditions.

This thesis is based on three papers on selected topics in Asymptotic Geometric Analysis. The first paper is about the volume of high-dimensional random polytopes; in particular, on polytopes generated by Gaussian random vectors. We consider the question of how many random vertices (or facets) should be sampled in order for such a polytope to capture significant volume. Various criteria for what exactly it means to capture significant volume are discussed. We also study similar problems for random polytopes generated by points on the Euclidean sphere. The second paper is about volume distribution in convex bodies. The first main result is about convex bodies that are (i) symmetric with respect to each of the coordinate hyperplanes and (ii) in isotropic position. We prove that most linear functionals acting on such bodies exhibit super-Gaussian tail-decay. Using known facts about the mean-width of such bodies, we then deduce strong lower bounds for the volume of certain caps. We also prove a converse statement. Namely, if an arbitrary isotropic convex body (not necessarily satisfying the symmetry assumption (i)) exhibits similar cap-behavior, then one can bound its mean-width. The third paper is about random polytopes generated by sampling points according to multiple log-concave probability measures. We prove related estimates for random determinants and give applications to several geometric inequalities; these include estimates on the volume-radius of random zonotopes and Hadamard's inequality for random matrices. / Mathematics

This study develops a multi-period log-return quantile forecasting procedure to evaluate the performance of eleven nearby commodity futures contracts (NCFC) using a sample of 897 daily price observations and at-the-money (ATM) put and call implied volatilities of the corresponding prices for the period from 1/16/2008 to 7/29/2011. The statistical approach employs dynamic log-returns quantile regression models to forecast price densities using implied volatilities (IVs) and factors estimated through principal component analysis (PCA) from the IVs, pooled IVs and lagged returns. Extensive in-sample and out-of-sample analyses are conducted, including assessment of excess trading returns, and evaluations of several combinations of quantiles, model specifications, and NCFC's. The results suggest that the IV-PCA-factors, particularly pooled return-IV-PCA-factors, improve quantile forecasting power relative to models using only individual IV information. The ratio of the put-IV to the call-IV is also found to improve quantile forecasting performance of log returns. Improvements in quantile forecasting performance are found to be better in the tails of the distribution than in the center. Trading performance based on quantile forecasts from the models above generated significant excess returns. Finally, the fact that the single IV forecasts were outperformed by their quantile regression (QR) counterparts suggests that the conditional distribution of the log-returns is not normal.

The world is experiencing yet another energy- and fuel predicament as oil prices are escalating to new hights. Alternative fuels are being promoted globally as the increasing gasoline prices trigger inflation. Basic food commodities are some of the goods hit by this inflation and the purpose of this thesis is to analyse whether the higher maize and sugar prices are having any effect on the expanding ethanol production. This thesis focuses on the two major crop inputs in ethanol production: maize (in the US) and sugar cane (in Brazil). Econometric tests using cross-sectional data were carried through to find the elasticities of the variables. The crops prices were tested against ethanol output using the log-linear model in several regressions to find a relationship. In addition, the output levels of the crops were tested using the same method. It was found that maize prices and output affects ethanol production. Sugar cane prices do not have any significant impact on ethanol production while sugar cane output has a small, yet significant relationhip with ethanol. Consequently, ethanol’s rise in the fuel market could be a result of increased maize input, rather than sugar. / Dagens värld upplever ännu ett energi- och bränsle predikament när oljepriser eskalerar mot nya höjder. Alternativa bränslen marknadsförs globalt samtidigt som de stigande bensinpriserna stimulerar inflationen. Några av de varor som drabbas av denna inflation är grundläggande livsmedelsprodukter och syftet med denna uppsats är att analysera huruvida de högre priserna på majs och socker påverkar den expanderande etanolproduktionen. Uppsatsen fokuserar på de två stora grödor som används som insatsvaror vid framställningen av etanol: majs (i USA) och sockerrör (i Brasilien). Ekonometriska tester genomfördes för att erhålla variablernas elasticiteter med hjälp av den cross-sectional data som behandlades. Genom log-linear modellen utfördes det ett antal regressioner för att hitta ett samband mellan grödornas priser och etanolproduktionen. Därutöver genomfördes tester för att hitta sambandet mellan grödornas utbud och etanol med hjälp av samma modell. Det upptäcktes att både pris och utbudet av majs påverkar etanolproduktionen. Sockerrörspriser har ingen signifikant inverkan på etanolproduktionen medan utbudet av sockerrör har en signifikant, om än svag, relation till etanol. Följaktligen kan etanols tillväxt i  bränslemarknaden tolkas som ett resultat av en stigande majsinsats snarare än sockerinstats vid etanolframställningen.

This thesis focuses on video watermarking robust against geometric attacks and video compressions. In addition to the requirements for an image watermarking algorithm, a digital video watermarking algorithm has to be robust against advanced video compressions, frame loss, frame swapping, aspect ratio change, frame rate change, intra- and inter-frame filtering, etc. Video compression, especially, the most efficient compression standard, H.264, and geometric attacks, such as rotation and cropping, frame aspect ratio change, and translation, are considered the most challenging attacks for video watermarking algorithms. In this thesis, we first review typical watermarking algorithms robust against geometric attacks and video compressions, and point out their advantages and disadvantages. Then, we propose our robust video watermarking algorithms against Rotation, Scaling and Translation (RST) attacks and MPEG-2 compression based on the logpolar mapping and the phase-only filtering method. Rotation or scaling transformation in the spatial domain results in vertical or horizontal shift in the log-polar mapping (LPM) of the magnitude of the Fourier spectrum of the target frame. Translation has no effect in this domain. This method is very robust to RST attacks and MPEG-2 compression. We also demonstrate that this method can be used as a RST parameters detector to work with other watermarking algorithms to improve their robustness to RST attacks. Furthermore, we propose a new video watermarking algorithm based on the 1D DFT (one-dimensional Discrete Fourier Transform) and 1D projection. This algorithm enhances the robustness to video compression and is able to resist the most advanced video compression, H.264. The 1D DFT for a video sequence along the temporal domain generates an ideal domain, in which the spatial information is still kept and the temporal information is obtained. With detailed analysis and calculation, we choose the frames with highest temporal frequencies to embed the fence-shaped watermark pattern in the Radon transform domain of the selected frames. The performance of the proposed algorithm is evaluated by video compression standards MPEG-2 and H.264; geometric attacks such as rotation, translation, and aspect-ratio changes; and other video processing. The most important advantages of this video watermarking algorithm are its simplicity, practicality and robustness.

This thesis focuses on video watermarking robust against geometric attacks and video compressions. In addition to the requirements for an image watermarking algorithm, a digital video watermarking algorithm has to be robust against advanced video compressions, frame loss, frame swapping, aspect ratio change, frame rate change, intra- and inter-frame filtering, etc. Video compression, especially, the most efficient compression standard, H.264, and geometric attacks, such as rotation and cropping, frame aspect ratio change, and translation, are considered the most challenging attacks for video watermarking algorithms. In this thesis, we first review typical watermarking algorithms robust against geometric attacks and video compressions, and point out their advantages and disadvantages. Then, we propose our robust video watermarking algorithms against Rotation, Scaling and Translation (RST) attacks and MPEG-2 compression based on the logpolar mapping and the phase-only filtering method. Rotation or scaling transformation in the spatial domain results in vertical or horizontal shift in the log-polar mapping (LPM) of the magnitude of the Fourier spectrum of the target frame. Translation has no effect in this domain. This method is very robust to RST attacks and MPEG-2 compression. We also demonstrate that this method can be used as a RST parameters detector to work with other watermarking algorithms to improve their robustness to RST attacks. Furthermore, we propose a new video watermarking algorithm based on the 1D DFT (one-dimensional Discrete Fourier Transform) and 1D projection. This algorithm enhances the robustness to video compression and is able to resist the most advanced video compression, H.264. The 1D DFT for a video sequence along the temporal domain generates an ideal domain, in which the spatial information is still kept and the temporal information is obtained. With detailed analysis and calculation, we choose the frames with highest temporal frequencies to embed the fence-shaped watermark pattern in the Radon transform domain of the selected frames. The performance of the proposed algorithm is evaluated by video compression standards MPEG-2 and H.264; geometric attacks such as rotation, translation, and aspect-ratio changes; and other video processing. The most important advantages of this video watermarking algorithm are its simplicity, practicality and robustness.

The purpose of this research was to quantify what value could be gained from cutting solid wood products from old-growth western hemlock (Tsuga heterophylla (Raf.) Sarg.) logs that are used to produce pulp in British Columbia. These logs represent a significant portion of the resource and increasing their value recovery would be beneficial to the forest industry. One hundred and sixteen logs were sampled from the coastal and interior regions of British Columbia. Dimension and quality attributes were measured to enable estimates of gross and merchantable volume. Logs deemed likely to yield lumber were sawn with the aim of maximizing value recovery. The nominal dimension and grade of all lumber recovered was recorded. Margins and breakpoints at which sawing became profitable were calculated. Models to predict the volume of lumber and proportion of Clear grade lumber recovered (“C Industrial” grade at the interior mill, “D Select” grade at the coastal mill) were developed. Lumber recovery, especially Clear grade lumber, was significantly higher from logs from the coastal site. At current market prices, cutting lumber from these logs was profitable, with the highest margins achieved when chips were produced from the milling residue. It was not profitable to recover lumber from the interior logs regardless of whether chips were produced. The disparity between locations was attributed to differences between the logs, the sawmilling equipment, the sawyers’ motivations and the lumber grades. Between 60% and 67% of coastal logs and 13% to 21% of interior logs returned a profit, depending on whether chips were produced. Models were developed to better identify these logs using observable attributes. A linear model described the total volume of lumber recovered. Significant predictor variables in the model were the gross log volume, the average width of the sound collar and the stage of butt/heart rot at the large end. A second model predicted the proportion of Clear grade lumber. Regional models were developed to account for different Clear lumber grades between sawmills. Significant predictor variables were knot frequency, diameter at the large end, volume, length, taper and the width of the sound collar at the large end.

In a message-flow Internet service where messages travel through multiple nodes, event log analysis is one of the most important methods to identify the root causes of problems. Traditional approaches for event log analysis have been largely based on expert systems that build static dependency models on rules and patterns defined by human experts. However, the semantic complexity and the various formats of event logs make it difficult to be modeled. In addition, it is time consuming to maintain such static model for constantly evolving Internet services. Recent research has been focused on building statistical models. However, all of these models rely on the trace information provided by J2EE or .NET frameworks, which are not available to all Internet services. In this thesis, we propose a framework of problem determination based on statistical analysis of event logs. We assume a unique message ID will be logged in multiple log lines to trace the message flow in the system. A generic log adaptor is defined to extract valuable information from the log entries. We also develop an algorithm of log event clustering and log pattern clustering. Frequency analysis will be performed based on the log patterns in order to build a statistical model of the system behaviors. Once the system is modeled, we can determine problems by running a chi-square goodness of fit test using a sliding window approach. As event logs are available on all major operating systems, we believe our framework is a generic solution for problem determination in message-flow Internet services. Our solution has been validated by the log data collected from the Blackberry Internet Service (BIS) engine ‎[4] , a wireless email service that serves millions of users across the world. According to the test results, our solution shows high accuracy of problem determination.

In a message-flow Internet service where messages travel through multiple nodes, event log analysis is one of the most important methods to identify the root causes of problems. Traditional approaches for event log analysis have been largely based on expert systems that build static dependency models on rules and patterns defined by human experts. However, the semantic complexity and the various formats of event logs make it difficult to be modeled. In addition, it is time consuming to maintain such static model for constantly evolving Internet services. Recent research has been focused on building statistical models. However, all of these models rely on the trace information provided by J2EE or .NET frameworks, which are not available to all Internet services. In this thesis, we propose a framework of problem determination based on statistical analysis of event logs. We assume a unique message ID will be logged in multiple log lines to trace the message flow in the system. A generic log adaptor is defined to extract valuable information from the log entries. We also develop an algorithm of log event clustering and log pattern clustering. Frequency analysis will be performed based on the log patterns in order to build a statistical model of the system behaviors. Once the system is modeled, we can determine problems by running a chi-square goodness of fit test using a sliding window approach. As event logs are available on all major operating systems, we believe our framework is a generic solution for problem determination in message-flow Internet services. Our solution has been validated by the log data collected from the Blackberry Internet Service (BIS) engine ‎[4] , a wireless email service that serves millions of users across the world. According to the test results, our solution shows high accuracy of problem determination.