The Frechet distribution was applied to a set of earthquake data in order to test its validity as a practical alternative distribution for extreme value data. It was concluded that the Frechet distribution was the best model representing that data set. Also, a Poisson model of occurrence could not be rejected for that data set. The combination of these two models resulted in a closed form unconditional extreme value distribution which was developed analytically. The appropriate statistical tests and sensitivity analyses were performed on the obtained model. / Applied Science, Faculty of / Mechanical Engineering, Department of / Graduate

The Subaru Coronagraphic Extreme Adaptive Optics (SCExAO) instrument, under development for the Subaru Telescope, has currently the fastest on-sky wavefront control loop, with a pyramid wavefront sensor running at 3.5 kHz. But even at that speed, we are still limited by low-frequency vibrations. The current main limitation was found to be vibrations attributed mainly to the rotation of the telescope. Using the fast wavefront sensors, cameras and accelerometers, we managed to identify the origin of most of the vibrations degrading our performance. Low-frequency vibrations are coming from the telescope drive in azimuth and elevation, as well as the elevation encoders when the target is at transit. Other vibrations were found at higher frequency coming from the image rotator inside Subaru's adaptive optics facility AO188. Different approaches are being implemented to take care of these issues. The PID control of the image rotator has been tuned to reduce their high-frequency contribution. We are working with the telescope team to tune the motor drives and reduce the impact of the elevation encoder. A Linear Quadratic Gaussian controller (LQG, or Kalman filter) is also being implemented inside SCExAO to control these vibrations. These solutions will not only improve significantly SCExAOs performance, but will also help all the other instruments on the Subaru Telescope, especially the ones behind A0188. Ultimately, this study will also help the development of the TMT, as these two telescopes share very similar drives.

The research question addressed by this study is: what is the relationship between Europeanization and the rise of extremist parties? In particular I examine the impact of Europeanization on the rise of extreme right parties in Europe from 1984 to 2006. Europeanization in this paper is defined as a process whereby the transformation of governance at the European level and European integration as a whole has caused distinctive changes in domestic politics. This process of Europeanization is one part of a structure of opportunities for extremist parties (which also include social, economic, and electoral factors). Although this study finds that Europeanization does not have a statistically significant effect it is still an important factor when examining domestic political phenomenon in Europe.

Participants in extreme sports have usually been studied from a risk perspective. The present study seeks to move away from this preconception and aims to study the emotional subjective experience. A qualitative methodology with phenomenological-hermeneutic design was used taking as a tool conversational systems. The informants were 8 extreme athletes belonging to the same group. Extreme sports included were: Mountain climbing, rappelling, bungee jumping, and mountaineering. The results indicate that fear is a generator of freedom, it can be useful, it is rewarding and it is also a promoter of personal transformations. This research gives theoretical value to the reason for involvement in extreme sports.

Extreme temperatures have increasingly played an important role in engineering applications, including leading edges during hypersonic flight, spacecraft re-entry, and propulsion systems. In order to design for such thermo-mechanical conditions, materials must be characterized using suitable measurement methods. DIC is a popular and versatile method in full-field measurement. In brief, DIC compares images of a sample between its undeformed and deformed state in order to get displacement and strain field maps. Since the images are acquired from digital cameras, it is important to have high contrast images for meaningful correlation. Exposure time is a pivotal camera setting relating to camera sensitivity. Alteration in exposure time results in variation of image contrast, thereby affecting DIC correlation. Also, it is well known that at extreme temperatures, materials emit light which can saturate DIC camera sensors, but the light can be mitigated using optical bandpass filters. In previous work, many have shown that blue bandpass filters can effectively extend the temperature range of DIC, and our lab has shown that ultraviolet (UV) filters can extend the range further. In this thesis, four different temperatures: room temperature, 1300°C, 1450°C, and 1600°C were tested by rigid-motion experiments. At each temperature level, UV images were acquired in order to examine the variation of DIC error over the whole range of exposure time. UV images were acquired at exposure times ranging from 500μs to 61,000μs, which are the minimum and maximum possible values for the cameras used in this thesis. The results showed that there were higher errors of UV-DIC at extremely dark or bright exposure times where as errors were generally insignificant at intermediate exposure times. In order to perform meaningful DIC up to 1600oC, the exposure time for the camera used in this thesis is suggested to be set between 10,000μs and 40,000μs.

In Canada, construction companies are facing disruptions to their operations due to bad or extreme weather conditions such as thunderstorms, heavy precipitation, flooding, heatwaves and snowstorms, which cause project delays, loss of productivity and increased financial costs. This sector is prone to more disruptions due to increase in the frequency, duration and intensity of extreme weather events due to future climate change. This study examined the impacts of extreme weather events on infrastructure development companies and investigated their current practices and actions to alleviate these impacts. A survey questionnaire was developed and administrated to owners, managers, engineers, supervisors and planners of construction companies. Apart from descriptive evaluations, the survey responses were quantitatively analyzed to determine the impact of bad weather conditions on the construction companies. The findings of this study suggested that most construction companies’ operations were delayed due to bad or extreme weather events. However, construction industry is not adopting proactive measures to avoid or minimize these impacts. The main environmental factors impacting construction companies, included flooding, high winds or thunderstorms, warm/cold temperatures, heatwaves and snow/ice storms. These bad weather impacts were more significant for non-government construction companies as compared to those working in the government sector. Indirect impacts of bad weather included disruptions to their supply chain networks and changes in customer behaviors; however, these impacts were minor compared to direct environmental impacts. The study found that both government and non-government sector construction companies granted accommodations to the workers during bad weather conditions; however, government sector companies were more accommodating as compared to non-government companies. The study results also provided insight into the financial impacts of extreme weather events on construction companies. Weighted average losses for government sector companies were $2,200 per day of bad weather as compared to $8,155 per day for non-government companies. This suggested that non-government construction companies may experience serious financial consequences due to bad or extreme weather events. Study results further showed that there were no adequate guidelines, protocols or standards available to construction companies to adapt their operations and planning for extreme weather events. The study also highlighted the lack of adequate insurance products available for the construction sector to deal with bad weather. There was little tendency shown by the construction companies to use new technologies to deal with bad weather conditions. Therefore, there is an urgent need to develop guidelines, protocols or standards for construction companies by involving all levels of the government and relevant private sector organizations. This study helps to determine the nature and scale of extreme weather impacts on construction industry and explores what strategies may be developed to alleviate these impacts and risks. Such knowledge will help companies better plan and manage their operations and effectively use their human resources. It will help in timely delivery of services and savings in costs by the infrastructure development companies, which are a major contributor to the Canadian economy. / Thesis / Master of Science (MSc)

Yes / This book is an edition of the Special Issue Advances in Modelling and Prediction on the Impact of Human Activities and Extreme Events on Environments that was published in Water journal.

Many real-world systems can be represented as network-graphs. Some of the networks have an inherent community structure based on interactions. The problem of identifying this grouping structure given a graph is termed as community detection problem which has certain existing algorithms. This thesis contributes by providing specific improvements to various community detection algorithms such as spectral clustering and extreme point algorithm. One of the main contributions is proposing a new sub-sampling method to make existing spectral clustering method scalable by reducing the computational complexity. Also, we have implemented extreme points algorithm for a general multiple communities detection case along with a sub-sampling based version to reduce the computational complexity. We have also developed spectral clustering algorithm for popularity-adjusted block model (PABM) model based graphs to make the algorithm exact thus improving its accuracy. / Master of Science / We live in an increasingly interconnected world, where agents constantly interact with each other. This general agent-interaction framework describes many important systems, such as social interpersonal systems, protein interaction systems, trade and financial systems, power grids, and the World Wide Web, to name a few. By denoting agents as nodes and their interconnections as links, any such system can be represented as a network. Such networks or graphs provide a powerful and universal representation for analyzing a wide variety of systems spanning a remarkable range of scientific disciplines. Networks act as conduits for many kinds of transmissions. For instance, they are influential in the dissemination of ideas, adoption of technologies, helping find jobs and spread of diseases. Thus networks play a critical role both in providing information and helping make decisions making them a crucial part of the Data and Decisions Destination Area. A well-known feature of many networks is community structure. Nodes in a network are often found to belong to groups or communities that exhibit similar behavior. The identification of this community structure, called community detection, is an important problem with many critical applications. For example, communities in a protein interaction network often correspond to functional groups. This thesis focuses on cutting-edge methods for community detection in networks. The main approach is efficient community detection via sub-sampling. This is applied to two different approaches. The first approach is optimization of a modularity function using a low-rank approximation for multiple communities. The second approach is a spectral clustering where we aim to formulate an algorithm for community detection by exploiting the eigenvectors of the network adjacency matrix.

Thesis (PhD (Statistics))--University of Stellenbosch, 2010. / ENGLISH ABSTRACT: In extreme value theory (EVT) the emphasis is on extreme (very small or very large) observations. The crucial parameter when making inferences about extreme quantiles, is called the extreme value index (EVI). This thesis concentrates on only the right tail of the underlying distribution (extremely large observations), and specifically situations where the EVI is assumed to be positive. A positive EVI indicates that the underlying distribution of the data has a heavy right tail, as is the case with, for example, insurance claims data. There are numerous areas of application of EVT, since there are a vast number of situations in which one would be interested in predicting extreme events accurately. Accurate prediction requires accurate estimation of the EVI, which has received ample attention in the literature from a theoretical as well as practical point of view. Countless estimators of the EVI exist in the literature, but the practitioner has little information on how these estimators compare. An extensive simulation study was designed and conducted to compare the performance of a wide range of estimators, over a wide range of sample sizes and distributions. A new procedure for the estimation of a positive EVI was developed, based on fitting the perturbed Pareto distribution (PPD) to observations above a threshold, using Bayesian methodology. Attention was also given to the development of a threshold selection technique. One of the major contributions of this thesis is a measure which quantifies the stability (or rather instability) of estimates across a range of thresholds. This measure can be used to objectively obtain the range of thresholds over which the estimates are most stable. It is this measure which is used for the purpose of threshold selection for the proposed PPD estimator. A case study of five insurance claims data sets illustrates how data sets can be analyzed in practice. It is shown to what extent discretion can/should be applied, as well as how different estimators can be used in a complementary fashion to give more insight into the nature of the data and the extreme tail of the underlying distribution. The analysis is carried out from the point of raw data, to the construction of tables which can be used directly to gauge the risk of the insurance portfolio over a given time frame. / AFRIKAANSE OPSOMMING: Die veld van ekstreemwaardeteorie (EVT) is bemoeid met ekstreme (baie klein of baie groot) waarnemings. Die parameter wat deurslaggewend is wanneer inferensies aangaande ekstreme kwantiele ter sprake is, is die sogenaamde ekstreemwaarde-indeks (EVI). Hierdie verhandeling konsentreer op slegs die regterstert van die onderliggende verdeling (baie groot waarnemings), en meer spesifiek, op situasies waar aanvaar word dat die EVI positief is. ’n Positiewe EVI dui aan dat die onderliggende verdeling ’n swaar regterstert het, wat byvoorbeeld die geval is by versekeringseis data. Daar is verskeie velde waar EVT toegepas word, aangesien daar ’n groot aantal situasies is waarin mens sou belangstel om ekstreme gebeurtenisse akkuraat te voorspel. Akkurate voorspelling vereis die akkurate beraming van die EVI, wat reeds ruim aandag in die literatuur geniet het, uit beide teoretiese en praktiese oogpunte. ’n Groot aantal beramers van die EVI bestaan in die literatuur, maar enige persoon wat die toepassing van EVT in die praktyk beoog, het min inligting oor hoe hierdie beramers met mekaar vergelyk. ’n Uitgebreide simulasiestudie is ontwerp en uitgevoer om die akkuraatheid van beraming van ’n groot verskeidenheid van beramers in die literatuur te vergelyk. Die studie sluit ’n groot verskeidenheid van steekproefgroottes en onderliggende verdelings in. ’n Nuwe prosedure vir die beraming van ’n positiewe EVI is ontwikkel, gebaseer op die passing van die gesteurde Pareto verdeling (PPD) aan waarnemings wat ’n gegewe drempel oorskrei, deur van Bayes tegnieke gebruik te maak. Aandag is ook geskenk aan die ontwikkeling van ’n drempelseleksiemetode. Een van die hoofbydraes van hierdie verhandeling is ’n maatstaf wat die stabiliteit (of eerder onstabiliteit) van beramings oor verskeie drempels kwantifiseer. Hierdie maatstaf bied ’n objektiewe manier om ’n gebied (versameling van drempelwaardes) te verkry waaroor die beramings die stabielste is. Dit is hierdie maatstaf wat gebruik word om drempelseleksie te doen in die geval van die PPD beramer. ’n Gevallestudie van vyf stelle data van versekeringseise demonstreer hoe data in die praktyk geanaliseer kan word. Daar word getoon tot watter mate diskresie toegepas kan/moet word, asook hoe verskillende beramers op ’n komplementêre wyse ingespan kan word om meer insig te verkry met betrekking tot die aard van die data en die stert van die onderliggende verdeling. Die analise word uitgevoer vanaf die punt waar slegs rou data beskikbaar is, tot op die punt waar tabelle saamgestel is wat direk gebruik kan word om die risiko van die versekeringsportefeulje te bepaal oor ’n gegewe periode.

The crude oil, natural gas, and electricity markets are among the most widely traded and talked about commodity markets across the world. Over the past two decades each commodity has seen price volatility due to political, economic, social, and technological reasons. With that comes a significant amount of risk that both corporations and governments must account for to ensure expected cash flows and to minimize losses. This thesis analyzes the portfolio risk of the major US commodity hubs for crude oil, natural gas and electricity by applying Extreme Value Theory to historical daily price returns between 2003 and 2013. The risk measures used to analyze risk are Value-at-Risk and Expected Shortfall, with these estimated by fitting the Generalized Pareto Distribution to the data using the peak-over-threshold method. We consider both the univariate and bivariate cases in order to determine the effects that price shocks within and across commodities will have in a mixed portfolio. The results show that electricity is the most volatile, and therefore most risky, commodity of the three markets considered for both positive and negative returns. In addition, we find that the univariate and bivariate results are statistically indistinguishable, leading to the conclusion that for the three markets analyzed during this period, price shocks in one commodity does not directly impact the volatility of another commodity’s price. / Graduate