Les progrès récents en sciences de la terre font de plus en plus de données multidisciplinaires disponibles pour l'exploration minière. Cela a permis le développement de méthodologies de calculer la prédictivité pour les zones aurifères basées sur des analyses statistiques des différents paramètres. L‘utilisation de nouveaux logiciels développés, la distribution spatiale et la topologie des polygones (e.g. intrusions granitiques) et des polylignes (e.g. zones de cisaillement) sont contrôlées par les paramètres définis par les utilisateurs (par exemple, la densité, la longueur, la surface, etc.). La distance des gisements d'or par rapport à polygones ou polylignes est calculée en utilisant une fonction de probabilité distribution. Les analyses statistiques des résultats de la modélisation montrent que i) les valeurs de la moyenne à la surface relative des polygones, la moyenne de longueur relative de polylignes, le nombre d'objets et leur regroupement sont essentiels à des évaluations statistiques ; ii) tester la validité des différents méthodes d'inversion dépend de l‘importance relative sur la corrélation entre les paramètres utilisés; et iii) la robustesse des points d'intérêt de distribution est de déduire des lois à l'égard de la qualité des données d'entrée. Cette approche a été appliquée aux données géologiques et géophysiques de l‘Yenissei Ridge d‘une surface totale de 75730 km2 pour cartographier la prédictivité de 29 nouvelles zones aurifères correspondant à une surface de 1811 km2. La méthode développée dans cette étude permet de réduire jusqu'à quatre fois de la superficie considérée comme des zones aurifères par des études précédentes. Pour la construction la plus précise des zones aurifères un modèle 3D de densité de l‘Yenissei Ridge a été construit. Ce modèle est basé sur les données de surface, gravimétriques et aéromagnétiques (grilles numérique de 1x1km2), profils sismiques et magnétotelluriques de "Batholite" et "Shpate". Ce modèle 3D de densité montre que: a) l‘Yenissei Ridge a une structure de couverture pliée, résultant d'un événement de la collision néoprotérozoïque ; b) γNPta granites Tatarsky-Ayhta et zones de cisaillement sont des seuls objets qui présente des relations spatiales avec la minéralisation aurifère. / Recent progresses in geosciences make more and more multidisciplinary data available for mining exploration. This allows developing methodologies to compute predictivity for gold zones by the statistical analysis of variable input parameters. Using newly developed software, the spatial distribution and the topology of polygons (e.g. intrusions) and polylines (e.g. shear zones) are controlled by parameters defined by users (e.g. density, length, surface, etc.). The distance of points of interest (gold deposits) with respect to a given type of objects (polygons or polylines) is given using a probability distribution function. The statistical analyses of output results from the direct modeling process show that i) values of relative surface mean of polygons, relative length mean of polylines, the number of objects and their clustering are critical to statistical appraisals, ii) the validity of the different tested inversion methods strongly depends on the relative importance and on the dependency between the parameters used, and iii) the robustness of the inferred distribution points of interest laws with respect to the quality of the input data. This approach was applied to the geological and geophysical data of the Yenissei ridge of the total area of 75730 km2 for the predictivity mapping of 29 new gold zones with the total area of 1811 km2. The newly developed method allows reducing up to four times of the area of predictivity gold zones, compared with previous studies. For more accurate construction of gold zones, a 3D density model of the Yenisei ridge was constructed. This model is based on surface gravity and aeromagnetic data (numerical grids of 1x1km), ―Batholite‖ and ―Shpate‖ seismic and magnetotelluric profiles, respectively. The 3D density model shows that: a) the Yenissei ridge has a cover-folded structure, formed during a Neopretorozoic collisional event, b) only γNPta Tatarsky-Ayhta granites and shear zones have spatial relationships with gold mineralization.

The hydrothermal vent environment, in its extreme spatial and temporal variability, offers the opportunity to study habitats that are naturally fragmented and unstable. The vestimentiferan tubeworm Ridgeia piscesae is a foundation species inhabiting hydrothermal vent habitat in the Northeast Pacific Ocean. R. piscesae is a phenotypically plastic species and is arranged in a metapopulation spatial structure, with each local population displaying one of a range of morphotypes. Ridgeia piscesae participates in an obligate symbiosis that is dependent on hydrogen sulphide in the hydrothermal vent fluid that supplies each local population. Hydrothermal fluid flow is highly variable in the hydrothermal vent environment and hydrogen sulphide flux is a limiting nutrient for R. piscesae; this variability may create differences in habitat quality. The objective of this study is to determine whether local populations of R. piscesae centered on high and low flux hydrothermal fluid outputs are similar in body condition, reproductive condition, and juvenile recruitment. Using the submersibles ROPOS and Alvin, I collected high flux and low flux sample pairs from within meters of each other at multiple sample sites on Axial Seamount and the Endeavour segment of the Juan de Fuca Ridge. I used morphological measurements, histology and lipid analysis to assess physiological and reproductive condition. I also determined the relative abundances of new and older recruits in high and low flux local populations. I found that low flux habitat was inferior in its ability to support Ridgeia piscesae at all stages in the tubeworm’s life cycle. In terms of body condition, local populations in low flux habitat had lower body weight, greater body length, smaller anterior tube diameter, lower trophosome volume, lower total lipid volume, and lower branchial plume condition. With respect to reproductive condition, local populations in low flux habitat had lower proportions of reproductive individuals, less sperm transfer, lower gonad volume, and fewer mature oocytes; there was no difference in sperm development stages between high and low flux habitat. From the perspective of the individual, low flux tubeworms live longer, and lifetime reproductive output may be comparable to high flux tubeworms. However, turnover is higher in the high flux habitat, so reproductive output of high flux populations is greater than that of low flux populations. Juvenile recruitment was biased toward high flux habitat, although this trend was not significant and recruitment to low flux habitat was still notable. The differences between reproductive output and juvenile recruitment between these habitats support a source-sink model of population dynamics. From the perspective of the metapopulation, low flux habitat is inferior in its ability to support Ridgeia piscesae at all stages in the tubeworm’s life cycle. This distribution of relative contributions to the overall population of a key species in a Marine Protected Area (MPA) should factor into management decisions affecting MPA boundaries and use. / Graduate

Thesis (M.A.)--University of Tennessee, 2007. / Title from title page screen (viewed on Mar. 31, 2008). Thesis advisor: Michael L. Keene. Vita. Includes bibliographical references.

Teilw. zugl.: @Diss. / Literaturverz. S. 253 - 267. Makoce ta wowasake: the power of the land -- Roots of ethnic difference -- Cattle, grass, and ethnic conflict at the grassroots -- The Oglala Omniciye and the struggle for land -- Doing their patriotic duty: the World War I takeover of the Oglala lands -- Representative democracy and the politics of exclusion -- Land and power in the era of the IRA -- A nation in crisis, poised for change.

Abstract Estimation of the covariance matrix is an important problem in statistics in general because the covariance matrix is an essential part of principal component analysis, statistical pattern recognition, multivariate regression and network exploration, just to mention but a few applications. Penalized likelihood methods are used when standard estimates cannot be computed. This is a common case when the number of explanatory variables is much larger compared to the sample size (high-dimensional case). An alternative ridge-type estimator for the precision matrix estimation is introduced in Article I. This estimate is derived using a penalized likelihood estimation method. Undirected networks, which are connected to penalized covariance and precision matrix estimation and some applications related to networks are also explored in this dissertation. In Article II novel statistical methods are used to infer population networks from discrete measurements of genetic data. More precisely, Least Absolute Shrinkage and Selection Operator, LASSO for short, is applied in neighborhood selection. This inferred network is used for more detailed inference of population structures. We illustrate how community detection can be a promising tool in population structure and admixture exploration of genetic data. In addition, in Article IV it is shown how the precision matrix estimator introduced in Article I can be used in graphical model selection via a multiple hypothesis testing procedure. Article III in this dissertation contains a review of current tools for practical graphical model selection and precision/covariance matrix estimation. The other three publications have detailed descriptions of the fundamental computational and mathematical results which create a basis for the methods presented in these articles. Each publication contains a collection of practical research questions where the novel methods can be applied. We hope that these applications will help readers to better understand the possible applications of the methods presented in this dissertation. / Tiivistelmä Kovarianssimatriisin estimointi on yleisesti ottaen tärkeä tilastotieteen ongelma, koska kovarianssimatriisi on oleellinen osa pääkomponenttianalyysia, tilastollista hahmontunnistusta, monimuuttujaregressiota ja verkkojen tutkimista, vain muutamia sovellutuksia mainitakseni. Sakotettuja suurimman uskottavuuden menetelmiä käytetään sellaisissa tilanteissa, joissa tavanomaisia estimaatteja ei voida laskea. Tämä on tyypillistä tilanteessa, jossa selittävien muuttujien lukumäärä on hyvin suuri verrattuna otoskokoon (englanninkielisessä kirjallisuudessa tämä tunnetaan nimellä ”high dimensional case”). Ensimmäisessä artikkelissa esitellään vaihtoehtoinen harjanne (ridge)-tyyppinen estimaattori tarkkuusmatriisin estimointiin. Tämä estimaatti on johdettu käyttäen sakotettua suurimman uskottavuuden estimointimenetelmää. Tässä väitöskirjassa käsitellään myös suuntaamattomia verkkoja, jotka liittyvät läheisesti sakotettuun kovarianssi- ja tarkkuusmatriisin estimointiin, sekä joitakin verkkoihin liittyviä sovelluksia. Toisessa artikkelissa käytetään uusia tilastotieteen menetelmiä populaatioverkon päättelyyn epäjatkuvista mittauksista. Tarkemmin sanottuna Lassoa (Least Absolute Shrinkage and Selection Operator) sovelletaan naapuruston valinnassa. Näin muodostettua verkkoa hyödynnetään tarkemmassa populaatiorakenteen tarkastelussa. Havainnollistamme, kuinka verkon kommuunien (communities) tunnistaminen saattaa olla lupaava tapa tutkia populaatiorakennetta ja populaation sekoittumista (admixture) geneettisestä datasta. Lisäksi neljännessä artikkelissa näytetään, kuinka ensimmäisessä artikkelissa esiteltyä tarkkuusmatriisin estimaattoria voidaan käyttää graafisessa mallinvalinnassa usean hypoteesin testauksen avulla. Tämän väitöskirjan kolmas artikkeli sisältää yleiskatsauksen tämänhetkisistä työkaluista, joiden avulla voidaan valita graafinen malli ja estimoida tarkkuus- sekä kovarianssimatriiseja. Muissa kolmessa julkaisussa on kuvailtu yksityiskohtaisesti olennaisia laskennallisista ja matemaattisista tuloksista, joihin artikkeleissa esitellyt estimointimenetelmät perustuvat. Jokaisessa julkaisussa on kokoelma käytännöllisiä tutkimuskysymyksiä, joihin voidaan soveltaa uusia estimointimenetelmiä. Toivomme, että nämä sovellukset auttavat lukijaa ymmärtämään paremmin tässä väitöskirjassa esiteltyjen menetelmien käyttömahdollisuuksia.

Submitted by Bruna Rodrigues (bruna92rodrigues@yahoo.com.br) on 2016-10-06T11:48:12Z No. of bitstreams: 1 DissMHC.pdf: 1077783 bytes, checksum: c81f777131e6de8fb219b8c34c4337df (MD5) / Approved for entry into archive by Marina Freitas (marinapf@ufscar.br) on 2016-10-20T13:58:41Z (GMT) No. of bitstreams: 1 DissMHC.pdf: 1077783 bytes, checksum: c81f777131e6de8fb219b8c34c4337df (MD5) / Approved for entry into archive by Marina Freitas (marinapf@ufscar.br) on 2016-10-20T13:58:47Z (GMT) No. of bitstreams: 1 DissMHC.pdf: 1077783 bytes, checksum: c81f777131e6de8fb219b8c34c4337df (MD5) / Made available in DSpace on 2016-10-20T13:58:52Z (GMT). No. of bitstreams: 1 DissMHC.pdf: 1077783 bytes, checksum: c81f777131e6de8fb219b8c34c4337df (MD5) Previous issue date: 2016-04-29 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / This paper presents a comparative study of the predictive power of four suitable regression methods for situations in which data, arranged in the planning matrix, are very poorly multicolinearity and / or high dimensionality, wherein the number of covariates is greater the number of observations. In this study, the methods discussed are: principal component regression, partial least squares regression, ridge regression and LASSO. The work includes simulations, wherein the predictive power of each of the techniques is evaluated for di erent scenarios de ned by the number of covariates, sample size and quantity and intensity ratios (e ects) signi cant, highlighting the main di erences between the methods and allowing for the creating a guide for the user to choose which method to use based on some prior knowledge that it may have. An application on real data (not simulated) is also addressed. / Este trabalho apresenta um estudo comparativo do poder de predi c~ao de quatro m etodos de regress~ao adequados para situa c~oes nas quais os dados, dispostos na matriz de planejamento, apresentam s erios problemas de multicolinearidade e/ou de alta dimensionalidade, em que o n umero de covari aveis e maior do que o n umero de observa c~oes. No presente trabalho, os m etodos abordados s~ao: regress~ao por componentes principais, regress~ao por m nimos quadrados parciais, regress~ao ridge e LASSO. O trabalho engloba simula c~oes, em que o poder preditivo de cada uma das t ecnicas e avaliado para diferentes cen arios de nidos por n umero de covari aveis, tamanho de amostra e quantidade e intensidade de coe cientes (efeitos) signi cativos, destacando as principais diferen cas entre os m etodos e possibilitando a cria c~ao de um guia para que o usu ario possa escolher qual metodologia usar com base em algum conhecimento pr evio que o mesmo possa ter. Uma aplica c~ao em dados reais (n~ao simulados) tamb em e abordada

La détection faible flux dans le moyen infrarouge (MIR) est fortement pénalisée par le rayonnement thermique de l’environnement. La principale solution à ce problème consiste à cryogéniser la plus grande partie possible de la chaîne de détection. Cette méthode atteint ses limites pour certaines applications,notamment l’imagerie haute résolution en astronomie par méthode interférométrique. Une solution alternative consiste à utiliser le processus non linéaire de somme de fréquences pour convertir ce rayonnement moyen infrarouge vers des domaines de longueur d’onde où les détecteurs ne sont plus limités par le rayonnement de l’environnement et fonctionnent efficacement en régime de comptage de photons. Les travaux effectués au cours ces trois années sont le prolongement direct de plus d’une décennie de recherche pour la détection et l’interférométrie par conversion de fréquence. Ils s’inscrivent dans la continuité des travaux de thèses de L. Szemendera et de P. Darré, ayant, pour l’un, posé les premières pierres de l’interférométrie par conversion de fréquence dans le MIR et, pour l’autre, démontré la possibilité d’utiliser cette technique sur le ciel à1550 nmen tirant parti de la technologie de guidage ridge. Cette thèse constitue donc la jonction fructueuse de ces travaux antérieurs, rendue possible par une collaboration avec l’institut Femto-ST. L’utilisation de leurs guides PPLN ridge nous a permis de nous placer au niveau de l’état de l’art pour la détection par conversion de fréquence dans le MIR (3,5 μm)avec des démonstrations expérimentales à la fois en laboratoire et sur le ciel (C2PU). Intégrée à un interféromètre à conversion de fréquence à ces mêmes longueurs d’onde, elle a également permis d’en améliorer significativement les performances et la répétabilité de cette technique, principalement dans le cas d’une source spectralement large bande. Parallèlement, notre collaboration avec le réseau de télescopes CHARA, nous a offert l’opportunité d’étudier les problématiques soulevées par une future mise en œuvre de l’instrument sur ce site. / Low flux detection in the mid infrared (MIR) is strongly penalized by thermal radiation from the environment. The main solution to this problem is to cryogenize as much of the detection chain as possible.This method is reaching its limits for some applications, including high-resolution imaging in astronomy using interferometric methods. An alternative solution is to use the non-linear process of sum frequency generation to convert this mid infrared radiation to wave length domains where detectors are no longer limited by the radiation from environment and operate efficiently in photon counting regime.The work over these three years is part of more than a decade of research into up-conversion detection and interferometry. It is the expansion of the these works of L. Szemendera and P. Darré. The first one laid the fondation of the up-conversion interferometry in the MIR and the second one demonstrated the possibility of using this technique on the sky at 1550 nm by benefiting from the ridge waveguide technology.Thus, this thesis constitutes the fruitful junction of these earlier works, made possible by a collaboration with the Femto-ST Institute. The use of their PPLN ridge waveguides has enabled us to place ourselves at the state of the art for up-conversion detection in the MIR (3,5 μm) with experimental demonstrations both in the laboratory and on the sky (C2PU). Integrated into a up-conversion interferometer at these same wave lengths, it has also significantly improved the performance and repeatability of this technique,in particular in the case of a spectrally broadband source. At the same time, our collaboration with the CHARA telescope array gave us the opportunity to study the problems raised by a future implementation of the instrument on this site.

No description available.

Thesis (Ph. D.)--Joint Program in Oceanography (Massachusetts Institute of Technology, Dept. of Earth, Atmospheric, and Planetary Sciences; and the Woods Hole Oceanographic Institution), 2013. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 160-163). / Physical oceanographers have known for several decades the total amount of abyssal mixing and upwelling required to balance the deep-water formation, but are still working to understand the mechanisms and locations-how and where it happens. From observational studies, we know that areas of rough topography are important and the hundreds of Grand-Canyon sized canyons that line mid-ocean ridges have particularly energetic mixing. To better understand the mechanisms by which rough topography translates into energetic currents and mixing, I studied diffusive boundary layers over varying topography using theoretical approaches and idealized numerical simulations using the ROMS model. In this dissertation, I show a variety of previously unidentified characteristics of diffusive boundary layers that are likely relevant for understanding the circulation of the abyssal ocean. These boundary layers share many important properties with observed flows in abyssal canyons, like increased kinetic energy near topographic sills and strong currents running from the abyssal plains up the slopes of the mid-ocean ridges toward their crests. They also have a previously unknown capacity to accelerate into overflows for a variety of oceanographically relevant shapes and sizes of topography. This acceleration happens without external forcing, meaning such overflows may be ubiquitous in the deep ocean. These boundary layers also can force exchange of large volumes of fluid between the relatively unstratified boundary layer and the stratified far-field fluid, altering the stratification far from the boundary. We see these effects in boundary layers in two- and three-dimensions, with and without rotation. In conclusion, these boundary layer processes, though previously neglected, may be a source of a dynamically important amount of abyssal upwelling, profoundly affecting predictions of the basin-scale circulation. This type of mechanism cannot be captured by the kind of mixing parameterizations used in current global climate models, based on a bottom roughness. Therefore, there is much work still to do to better understand how these boundary layers behave in more realistic contexts and how we might incorporate that understanding into climate models. / by Rebecca Walsh Dell. / Ph.D.

Inom fondförvaltning används kvantitativa metoder för att förutsäga hur företags räkenskaper kommer att förändras vid nästa kvartal jämfört med motsvarande kvartal året innan. Banken SEB använder i dag multipel linjär regression med förändring av intäkter som beroende variabel och förändring av valutakurser som oberoende variabler. Det är problematiskt av tre anledningar. Först och främst har valutor ofta stor multikolinjäritet, vilket ger instabila skattningar. För det andra det kan ett företags intäkter bero på ett urval av de valutor som används som data varför regression inte bör ske mot alla valutor. För det tredje är nyare data mer relevant för prediktioner. Dessa problem kan hanteras genom att använda regulariserings- och urvalsmetoder, mer specifikt elastic net och viktad regression. Vi utvärderar dessa metoder för en stor mängd företag genom att jämföra medelabsolutfelet mellan multipel linjär regression och regulariserad linjär regression med viktning. Utvärderingen visar att en sådan modell presterar bättre i 65,0 % av de företag som ingår i ett stort globalt aktieindex samt får ett medelabsolutfel på 14 procentenheter. Slutsatsen blir att elastic net och viktad regression adresserar problemen med den ursprungliga modellen och kan användas för bättre förutsägelser av intäkternas beroende av valutakurser. / Quantative methods are used in fund management to predict the change in companies' revenues at the next quarterly report compared to the corresponding quarter the year before. The Swedish bank SEB already uses multiple linear regression with change of revenue as the depedent variable and change of exchange rates as independent variables. This is problematic for three reasons. Firstly, currencies often exibit large multicolinearity, which yields volatile estimates. Secondly, a company's revenue can depend on a subset of the currencies included in the dataset. With the multicolinearity in mind, it is benifical to not regress against all the currencies. Thirdly, newer data is more relevant for the predictions. These issues can be handled by using regularisation and selection methods, more specifically elastic net and weighted regression. We evaluate these methods for a large number of companies by comparing the mean absolute error between multiple linear regression and regularised linear regression with weighting. The evaluation shows that such model performs better for 65.0% of the companies included in a large global share index with a mean absolute error of 14 percentage points. The conclusion is that elastic net and weighted regression address the problems with the original model and can be used for better predictions of how the revenues depend on exchange rates.