Dans sa première partie, cette thèse étudie l’utilisation optimale des produits dérivés par les intermédiaires financiers dans leur gestion du risque, en prêtant spécifiquement attention au marché des dérivés de taux d’intérêt. En modélisant la structure de capital optimale d’une banque, le premier chapitre montre comment l’usage optimal des dérivés affecte certaines décisions souvent étudiées en finance d’entreprise : l’offre de crédit, la transformation de maturité, la politique de dividendes ou les probabilités de défaut. La seconde partie de la thèse étudie au contraire le marché des dérivés comme un système à part entière. Le second chapitre utilise une base de données nouvelle et unique d’expositions bilatérales sur des contrats CDS afin d’offrir une description détaillée de la structure du réseau des expositions. Le troisième chapitre a pour objet la régulation des marchés de produits dérivés. Il étudie la compensation centrale des produits dérivés standardisés, et la demande de collatéral induite par cette réforme à l’échelle mondiale, sous une variété d’hypothèses concernant la microstructure du marché. / In its first part, this thesis studies the optimal use of derivatives contracts for risk management by financial intermediaries, focusing especially on interest rate derivative contracts. It models the optimal capital structure policy of a bank and shows how the optimal use of derivatives affects a number of oft-studied decisions in corporate finance: bank lending, maturity mismatching, payout policy or default probabilities. The second part of the thesis, in contrast, studies derivatives market as a system on its own. The second chapter uses a new and unique dataset of bilateral exposures to CDS contracts in order to provide a detailed description of the network structure of exposures. The third chapter focuses on the regulation of derivatives markets. It studies central clearing of standardized derivatives contracts and the collateral demand induced by the reform at a global scale, under a variety of hypotheses regarding the market microstructure.

The topic of this master’s thesis is design of ride-on scrubber-dryer. The thesis concerns own design, which is based on findings from the analysis of the current state of knowledge.

MENVSC / Department of Hydrology and Water Resources / Africa is endowed with various earthy materials, of which each community is uniquely identified. Inspite of the vast knowledge of western cosmetics, application of traditionally used cosmetics have been in existence and will be into practice from generation to generation, this is as a result of their various indigenous knowledge about traditional cosmetics. They are mostly applied during traditional rites and festive periods. Occasionally, earthy materials play a vital role in traditionally used cosmetics. Different reasons have been associated with the use of these earthy materials, they are: for skin cleansing, protection against ultra violet radiation, skin lightening, perfecting the distorted part of the skin and improving beauty. This study is focused on the physico-chemical, chemical and mineralogical characterisation of earthy materials indigenously used as cosmetics. This is achieved firstly by identifying the various earthy materials applied topically as well as their constituents. Samples for this study were obtained from six (6) different African countries namely: Botswana, Cameroon, Democratic Republic of the Congo, Nigeria, South Africa and Swaziland and were subjected to the following analyses: Scanning Electron Microscope and Electron Dispersive X-ray Spectroscopy for the morphology and elemental analysis, hydrometer method for the determination of particle sizes of each sample and a texture Auto Lookup Software Package (TAL Version 4.2) for the classification of their various textures (%clay, %sand and %silt). The X-ray diffractometry was applied for the identification of their mineral phases (secondary, major and minor). The Brunauer Emmett and Teller (BET) method was used for the physisorption analysis and classifying them according to their isotherm types (Type I, II, III, IV, V and VI). The chemical analysis was achieved by X-ray fluorescence (XRF). The result revealed that the morphology of the various representative samples varies, except CMR 1; 2, SCC 1; 2 and SLW 1; 2 which showed same structure and same elemental constituents, their resemblance indicated that they possess same geological formation. The XRF result indicated the presence of the following trace elements: Al, As, Au, Ba, Bi, Br, Ca, Cd, Ce, Cl, Co, Cr, Cs, Dy, Er, Eu, Fe, Ga, Gd, Ge, Hf, Hg, Ho, I, In, K, La, Lr, Lu, Mg, Mn, Mo, Na, Nd, Ni, Nb, Os, P, Pb, Pd, Pr, Pt, Re, Rh, Ru, S, Sc, Se, Sb, Si, Sm, Sn, Sr, Ta, Te, Tb, Th, Ti, U, V, W, Y, Yb, Zr and Zn, were measured in ppm. For example, Fe – 66.47 (Qwa-Qwa), Al – 4.64 (Swaziland), Si – 6.26 (Democratic Republic of Congo), Ti – 3.75 (Qwa-Qwa) and so on. From the particle size distribution analysis, some samples were identified to be clay (BTS 1, BTS 2, NG 1, SCC 4, SCC 5, SCC 9, SCC 11, SCC 12, SWL 1 and SWL 2) with their particle sizes < 2μm, whereas other samples showed other properties which are > 2 μm. Ten minerals were identified in the clay fraction samples: albite, goethite, halloysite, hematite, kaolinite (kaolinite-1Md, kaolinite 1) Mica (muscovite), sulphur, talc, chalcanthite and whewellite. The BET method vii for physical adsorption revealed that BTS 1, BTS 2, CMR 1, CMR 2, DRC A, DRC B, NG 1, NG 2, SWL 1, SWL 2, SCC 1, SCC 4, SCC 5, SCC 8, SCC 9, SCC 11 and SCC 12 countries possess Type II isotherm whereas samples from SCC 2, SCC 6, SCC 7 and SCC 10 had Type III isotherm. The physisorption analysis result also determined the various surface areas and porosity of each representative samples.

Within the banking industry many tasks, that previousrelied upon by manual labour is continuously being exchangedfor automated solutions. With the update in themethod of performing services through advanced software,benefits such as decrease in cost, increase in handling capacityor decrease in management delay come into focus.In the sector of Clearing Services the automation of simulatingrisk and clearing trades is being less of a questionand more of a requirement, since an increase in demand forderivatives clearing is expected. With this as a focus pointthis research project studied the potential effects that achange from manual to automation might present. To alsograsp a better understanding of the future market and competitivepotential the current situation on the market wasalso researched, where regulation proved itself as one ofthe motivating factors behind an increased activity. Thestudy concluded not only a decrease in cost, handling delayand increase in clearing capacity, but also future requirementfor implementation if any sustainable positionamongst market competition is to be expected. In additionthe resources that are relieved from the performing simulationsand take-ups would be able to focus on projects thatwould increase product value for the customer. / Inom bankindustrin overgår i dagsläget en stor andelprocesser från manuell hantering till att bli automatiserade.Genom uppdateringen till avancerad mjukvara, i relationtill hur tjänster hanteras och genomfors, så framkommerfordelar såsom exempelvis reduktioner i kostnader, processfordrojning,samt okad hanteringskapacitet. Automatiseringi sektorn Clearing Services är mer eller mindre ettkrav, då okad efterfråga utav clearing av derivat är forväntad.Med detta som fokus har denna studie undersoktpotentiella effekter som kan framkomma i en overgång frånmanuellt arbete till automatisering. For att bättre forståden framtida marknaden och den potentiella konkurrenskraften,undersokdes även den nuvarande marknaden. Meddetta identifierades även den motiverande faktorn bakomregulering och hur den påverkar det okande intresset gällandeclearing av derivat. Studien nådde fram till slutsatsenatt implementation skulle leda till reduktioner av kostnader,processfordrojning och okad hanteringskapacitet. Foratt behålla en kompetitiv position på marknaden for derivatclearing så kommer automatisering att krävas. Vidareskulle frilagda resurser från automatisering även kunna användasfor att utveckla produkter vilka okar kundvärdet.

Within Central Counterparty Clearing, the Clearing House collects Initial Margin from its Clearing Members. The Initial Margin can be calculated in many ways, one of which is by applying the commonly used risk measure Value-at-Risk. However, Value-at-Risk has one major flaw, namely its inability to encapsulate Tail Risk. Due to this, there has for long been a desire to replace Value-at-Risk with Expected Shortfall, another risk measure that has shown to be much better suited to encapsulate Tail Risk. That said, Value-at-Risk is still used over Expected Shortfall, something which is mainly due to the fact that there is no consensus regarding how one should backtest Expected Shortfall. The goal of this thesis is to evaluate some of the most commonly proposed methods for backtesting Expected Shortfall. In doing this, several non-parametric backtests of Expected Shortfall are investigated using simulated data as well as market data from different types of securities. Moreover, this thesis aims to shed some light on the differences between Value-at-Risk and Expected Shortfall, highlighting why a change of risk measure is not as straightforward as one might believe. From the investigations of the thesis, several backtests are found to be sufficient for backtesting the Initial Margin with Expected Shortfall as the risk measure, the so called Minimally Biased Relative backtest showing the overall best performance of the looked at backtests. Further, the thesis visualizes how Value-at-Risk and Expected Shortfall are two risk measures that are inherently different in a real-world setting, emphasizing how one should be careful making conversions between the two based upon parametric assumptions.

The encroachment of rangelands by woody plants causes an imbalance in the grass:bush ratio leading to decline in ecosystem services including grazing capacity, biodiversity, and water yield. This study assessed the historical changes in woody cover, and the effects of bush clearing on water loss, atmospheric carbon uptake, herbaceous vegetation, and soil seed bank (SSB) dynamics. The study was conducted at Makapanstad (Radi and Maseding) and Kgomokgomo rangelands, North-West Province, South Africa. Landsat imagery was used to assess woody cover over 34 years (1984 - 2018) in a total area of 16 397 ha. Evapotranspiration (ET) and gross primary productivity (GPP) were assessed in cleared sites and their adjacent uncleared sites from 2013 to 2018. Woody plant densities, herbaceous composition, ground cover and biomass production were assessed in 24, 5×5-m plots distributed equally in three 2500-m2 bush-encroached blocks in each rangeland in February 2016. Three samples for soil nutrient analysis were collected 20-cm deep in each plot. Thereafter, half (1250-m2) of each block was mechanically cleared to make three replicates of cleared and uncleared microsites. Post-treatment survey of vegetation was conducted in February 2017. SSB was assessed for samples collected in April 2016 before bush clearing and in August, October and December 2016, and April and August 2017 in paired microsites. In 1984, herbaceous cover (humid + arid grasslands) dominated, accounting for 36% of the total area followed by woody cover (shrublands + bushlands) and unvegetated cover (bare soils + abandoned croplands), with respective proportions of 33 and 31%. There were substantial land cover changes over time, with shrub cover increasing linearly (r2 = 0.94, p < 0.05) at a rate of 0.26% year-1, increasing the total woody cover (TWC) to 38% in 2018. Cropland abandonment was the main driver of increase in shrub cover in the study area. Herbaceous cover declined with increase in TWC (r2 = 0.69), whereas bare soil cover increased (r2 = 0.70) with TWC over time. The main encroachers were Vachellia tenuispina at Radi, Vachellia tortilis at Maseding and Dichrostachys cinerea and V. tortilis at Kgomokgomo. Most of the woody species correlated with total N and clay content (r = 0.96; p < 0.05) in black vertic clay soils of Radi and Maseding, whereas non-leguminous species correlated with total N and C:N in the red-yellow apedal soils of Kgomokgomo. ET and GPP varied interannually in all rangelands and both declined significantly (p < 0.05) after bush clearing at Radi and Kgomokgomo but not at Maseding. ET increased with GPP in cleared (r2 = 0.50-0.59) and uncleared sites (r2 = 0.82-0.93) at Radi and they showed a strong relationship (r2 >0.70) in both sites at Maseding and Kgomokgomo. ET ranged from 0.26-0.46 and 0.48-0.97 Kg H2O m-2 day-1 in cleared and uncleared sites, respectively, at Radi and from 0.44-0.98 and 0.63-1.24 Kg H2O m-2 day-1 at Kgomokgomo. Total SSB densities (constituted mostly by early successional species) increased significantly (p < 0.05) to 2470, 1872 and 693 seeds m-2 at Radi, Maseding and Kgomokgomo, respectively, in cleared microsites in August 2017. Similarly, biomass production (BP) and basal cover (BC) were highest in cleared relative to uncleared microsites. Overall, increase in woody cover reduced herbaceous cover and this calls for bush control in the studied rangelands. The decline in ET after bush clearing suggested that bush clearing could be used to increase rangeland water yield. Increased SSB densities, BP and BC indicated that bush clearing facilitates passive restoration. / Thesis (PhD (Pasture Science))--University of Pretoria, 2020. / Agricultural Research Council / National Research Foundation / Plant Production and Soil Science / PhD (Pasture Science) / Unrestricted

Machine learning is a rapidly growing field within artificial intelligence that an increasing number of individuals and corporations are beginning to utilize. In recent times, the financial sector has also started to recognize the potential of these techniques and methods. Nasdaq Clearing is responsible for managing the clearing business for the clearinghouse's members, and the objective of this thesis has been to explore the possibilities of using machine learning to replicate a subpart of the SPAN model's margin call calculations, known as initial margin, in the commodities market. The purpose of replicating SPAN's initial margin calculations is to open up for possibilities to create transparency and understanding in how the input variables affect the output. In the long run, we hope to broaden the insights on how one can use machine learning within the margin call processes. Various machine learning algorithms, primarily focused on regression tasks but also a few classification ones, have been employed to replicate the initial margin size. The primary objective of the methodology was to determine the algorithm that demonstrated the best performance in obtaining values that were as close as possible to the actual initial margin values. The findings revealed that a model composed of a combination of classification and regression, with non-parametric algorithms such as Random Forest and KNN, performed the best in both cases. Our conclusion is that the developed model possesses the ability to effectively compute the size of the initial margin and thus accomplishes its objective. / Maskininlärning är ett snabbt växande område inom artificiell intelligens som allt fler individer och företag börjar använda. Finanssektorn har nu också börjat undersöka hur dessa tekniker och metoder kan skapa värde. Nasdaq Clearing hanterar clearingverksamheten för clearinghusets medlemmar och syftet med denna uppsats har varit att undersöka möjligheterna att använda maskininlärning för att replikera en del av SPAN-modellens beräkningar av marginkravet som kallas Initial Marginal. Syftet med att replikera SPANs initiala marginberäkningar är att öppna upp för möjligheter att skapa transparens och förståelse för hur inputvariablernapåverkar outputen. På sikt hoppas vi kunna bredda insikterna hur maskininlärningslösningar skulle kunna användas inom "Margin Call"- processen. De metoder som användes för att replikera storleken på Initial Margin var olika maskininlärningsalgoritmer, främst fokuserade på regressionsuppgifter men några klassificeringsalgoritmer användes också. Fokus i metoden var att hitta vilken algoritm som presterade bäst, det vill säga den algoritm som predikterade närmst de faktiska värdena för Initial Margin. Resultatet visade sig vara en modell som kombinerade klassificering och regression, där icke-parametriska algoritmer såsom Random Forest och KNN var de som presterade bäst i båda fallen. Vår slutsats är att den utvecklade modellen har en god förmåga att beräkna storleken på Initial Margin och därmed uppfyller den sitt syfte.

The female reproductive organs are among the few adult tissues in which periodic angiogenesis normally occurs. Pathological angiogenesis can occur in various conditions, such as solid tumors. Vascular endothelial growth factor (VEGF) signaling often represents a critical rate-limiting step in physiological and pathological angiogenesis. This study utilizes development of utero-ovarian vasculature during pregnancy in goats as a model of physiological angiogenesis. Non-pregnant does and does at 4, 7, 10, 13, 16, and 18 weeks of gestation were used. Arteries of the reproductive tract were injected <i>in situ</i> with Microfil®. The tracts were fixed, dehydrated, and rendered transparent to reveal the paths of arteries. The ovarian artery was tortuous and lay in close apposition to the uterine tributary of the ovarian vein in all specimens studied. In non-pregnant does, this arrangement may serve as a local utero-ovarian pathway for the corpus luteum (CL) luteolysis at the end of non-fertile estrous cycle. During pregnancy, this arterio-venous arrangement may transfer luteotropic substances from uterus to ovary, which may serve in maternal recognition of pregnancy and fit the fact that the goat is CL dependent throughout gestation. In some cases of triplets, the size of the uterine branch of the ovarian artery was equal to or even larger than that of its parent artery and/or the ipsilateral uterine artery; and the vaginal artery contributed a connecting branch to the uterine artery. These physiological adaptations of the ovarian and/or vaginal arteries correlate well with the increasing nutrient demands of the growing multiple fetuses. In a second experiment, the vasculature of the uterus and ovaries was injected <i>in situ</i> with a mixture of Batson's No.17® and methyl methacrylate and then processed for observation by SEM. The microvasculature differed between non-pregnant and pregnant does, and with advancing gestation. We concluded that goats possess a <i>multivillous</i> type placenta. Capillary sinusoids and crypts on the fetal surface of the caruncle may compensate for the negative effect of the increased interhemal distance. Intussusceptive angiogenesis should be considered as equally possible and important mechanism as sprouting angiogenesis during placental development. Capillary diameters increased significantly during pregnancy especially after 4 weeks. Capillary density index was 66.8, 68.7, 55.5, 63.5, 70.1, 70.4, 64.5 percent in non-pregnant, 4, 7, 10, 13, 16, and 18 weeks of pregnancy, respectively. In the ovary, coiling of the ovarian branch of the ovarian artery around the ovarian tributary of the ovarian vein was observed. This may represent a local channel required for product transport from ovarian vein to ovarian artery and might have a role in regulating blood pressure to various ovarian structures. Immunolocalization of VEGF was performed as a third experiment. Immunostaining was observed in cyto- trophoblasts, maternal epithelial tissues, and vascular endothelium and smooth muscle, but not in binucleate giant cells or connective tissue. No apparent differences were observed in intensity and pattern of VEGF staining associated with advancing gestation. Luteal and follicular cells, and endothelium and smooth muscles of the ovarian vasculature positively stained. Patterns and intensity of staining of VEGF suggest that the fetus is directing its own survival by producing growth factors affecting fetal and maternal tissues. VEGF may have a role in growth and differentiation of cytotrophoblasts, as well as, development and maintenance of CL. In the fourth experiment, the sequential expression of VEGF and its receptors (fms-like tyrosine kinase, Flt-1 and kinase-insert domain-containing receptor, KDR) was measured using real-time quantitative PCR. Targets were detected in all studied tissues; however, levels of expression differed according to the stage of pregnancy. Expression of VEGF and its receptor mRNAs increased with advancing pregnancy, which correlates with the expansion of vasculature during pregnancy. Differences in the time-courses of the expression of Flt-1 and KDR mRNAs during pregnancy suggest that each receptor plays a different role in the angiogenic process. As an application of our model of angiogenesis, we tested the effect of swainsonine (active compound of locoweed and a potential anti-cancer drug) on the process. Does treated with swainsonine were euthanized at 7 and 18 weeks. No significant differences were found in sinusoidal diameters in treated does at 7 weeks, but a decrease in capillary density index was noted. In the ovary, focal avascular areas were observed in the corpus luteum of swainsonine-treated does at 7 weeks of pregnancy. Swainsonine caused great distortion in the uterine and ovarian vasculature at 18 weeks. A decrease in intensity of the immunoreactivity to VEGF antibody was observed in tissues from swainsonine-treated does at 7 and 18 weeks. There was no substantial effect of swainsonine on the expression of VEGF and its receptors' mRNAs in any of the studied tissues (except in the left ovary, where it had an inhibitory effect) at 7 weeks of pregnancy, but it had an inhibitory effect at 18 weeks. Demonstration of swainsonine's potential to negatively affect vascular development and suppress genes likely involved in angiogenesis at critical stages of blood vessel proliferation lends credibility to its potential as anti-cancer drug. / Ph. D.

The main objective of this project was to determine the possibility and feasibility of creating wildlife clearings through the use of herbicides as compared to conventional methods of bulldozing and hand clearing. The selected study area adjoining a series of 3 compartments designated as Broad Run Research Area, designed to study various methods of habitat manipulation. Twelve areas were selected for treatment and numbered consecutively. Two were abandoned; No. 1 because of its small size and No. 3 because of the value in its natural condition. A square clearing design was arranged in each of the remaining 10 areas. Four herbicides were selected for testing, each proposed clearing was divided into 4 equal sections in order to apply the 4 herbicides to each clearing; one per quarter-section. The 4 herbicides tested were Monuron, Fenuron, 2,4,5-T, and Ammonium sulfamate. Nonuron and Fenuron, in granular form, were applied around the base of each stem at rates of 5 and 10 grams per stem. Stems 0-4.9 inches d.b.h. received 5 grams and stems 5 inches d.b.h. and larger received 10 grams. Ammonium sulfamate and 2,4,5-T were applied to frills cut at waist height. The ax cute in 2,4,5-T treatment were spaced at 4 inch intervals, and in the ammonium sulfamate treatment were continuous and overlapping. These two chemicals were applied by the use of 2 gallon garden sprayers; ammonium sulfamate at a rate of 7lbs. of crystals dissolved in 2 gallons of water and 2,4,5-T at a rate of 12 lbs. active ingredient per 100 gals. of No. 2 diesel fuel. Frills were filled to overflowing. The granular herbicides were applied in June and July 1958 and the frill treatments were made in August of that year. Two stem counts by species were made on each entire area. The first was made immediately after treatment and the second after one complete growing season. The percent of kill was computed for each herbicide based on the stem counts. Herbaceous sample plots 1/100 acre in size, were established in 4 areas representing 4 topographical features. Two stem counts were made on each of these 4 areas. An incidental field study using 1,2, and 4-grams of Fenuron per stem revealed that most woody species can be controlled with 1 gram of active ingredient per stem. Openings were satisfactorily established with Fenuron and Monuron; a lower rate of application could have been used. Techniques used in frilling would have to be altered to obtain satisfactory results when using 2,4,5-T and ammonium sulfamate; complete frills are necessary for 2,4,5-T treatments and the ax cuts placed closer to the root collar in ammonium sulfamate treatments. Sight observations revealed the areas were being used by deer, rabbit, grouse, turkey, and woodcock. / Master of Science

The 2007 financial crisis highlighted the severe risks posed by counterparty defaults in financial markets. Assessing and addressing counterparty credit risk has consequently been a focal point of new regulations introduced in the wake of the crisis. The Central Clearing Counterparty (CCP) is at the heart of the solution, an entity dedicated to managing and mitigating counterparty risk in a market. CPPs manage risk by collecting collateral, referred to as margin, from the participants trading on the market. Appropriately sizing the margin is of utmost importance for the CCP to maintain the integrity of its operation and, by extension, protect the participants in the market. Most contemporary margin methodologies require significant resources which precludes frequent margin updates. In light of this issue, our work examines the capability of replicating the popular margin methodology Historical Simulation Value at Risk using machine-learning-based methods envisioning that an adequate such model could be used as a complement to the traditional model, providing real-time margin estimations. The experiment concerns portfolios containing stocks, bonds, and options and uses static market data and scenarios. We conclude that neither of the ensemble methods are sufficiently accurate, while both of the neural network-based models show moderate promise, warranting further development.