Approved for public release; distribution is unlimited. / This paper documents the development of a statistical analysis package for the TRS-80 microcoraputer. The package is comprised of six interactive programs which are generally divided into topical areas. The major emphasis is on exploratory data analysis and statistical inference, however, probability and inverse probability distributions are also included. The programming language is TRS-80 Level II BASIC enhanced by the input/output commands available through the ESF-80 (Exatron Stringy Floppy) mass storage subsystem. With the modification of these few commands, the package is compatible with most floppy disk operating systems designed for the TRS-80 Model I or Model III microcomputers. This statistical analysis capability implemented on a relatively inexpensive system provides a useful tool to the student or the trained analyst without ready access to a mainframe computer system. / Major, United States Marine Corps

A modelagem da volatilidade desempenha um papel fundamental em Econometria. Nesta dissertação são estudados a generalização dos modelos autorregressivos condicionalmente heterocedásticos conhecidos como GARCH e sua principal generalização multivariada, os modelos DCC-GARCH (Dynamic Condicional Correlation GARCH). Para os erros desses modelos são consideradas distribuições de probabilidade possivelmente assimétricas e leptocúrticas, sendo essas parametrizadas em função da assimetria e do peso nas caudas, necessitando assim de estimar esses parâmetros adicionais aos modelos. A estimação dos parâmetros dos modelos é feita sob a abordagem Bayesiana e devido às complexidades destes modelos, métodos computacionais baseados em simulações de Monte Carlo via Cadeias de Markov (MCMC) são utilizados. Para obter maior eficiência computacional os algoritmos de simulação da distribuição a posteriori dos parâmetros são implementados em linguagem de baixo nível. Por fim, a proposta de modelagem e estimação é exemplificada com dois conjuntos de dados reais / The modeling of volatility plays a fundamental role in Econometrics. In this dissertation are studied the generalization of known autoregressive conditionally heteroscedastic (GARCH) models and its main principal multivariate generalization, the DCCGARCH (Dynamic Conditional Correlation GARCH) models. For the errors of these models are considered distribution of probability possibility asymmetric and leptokurtic, these being parameterized as a function of asymmetry and the weight on the tails, thus requiring estimate the models additional parameters. The estimation of parameters is made under the Bayesian approach and due to the complexities of these models, methods computer-based simulations Monte Carlo Markov Chain (MCMC) are used. For more computational efficiency of simulation algorithms of posterior distribution of the parameters are implemented in low-level language. Finally, the proposed modeling and estimation is illustrated with two real data sets

The electron emission characteristics of aluminum, molybdenum and carbon nanotubes were studied. The experiments were setup to study the emission behavior as a function of temperature and exposure to oxygen. Changes in the surface work function as a result of thermal annealing were monitored with low energy ultra-violet photoelectron spectroscopy for flat samples while field emission energy distributions were used on tip samples. The change in the field emission from fabricated single tips exposed to oxygen while in operation was measured using simultaneous Fowler-Nordheim plots and electron energy distributions. From the results a mechanism for the degradation in the emission was concluded. Thermal experiments on molybdenum and aluminum showed that these two materials can be reduced at elevated temperatures, while carbon nanotubes on the other hand show effects of oxidation. To purely reduce molybdenum a temperature in excess of 750 ºC is required. This temperature exceeds that allowed by current display device technology. Aluminum on the other hand shows reduction at a much lower temperature of at least 125 ºC; however, its extreme reactivity towards oxygen containing species produces re-oxidation. It is believed that this reduction is due to the outward diffusion of aluminum atoms through the oxide. Carbon nanotubes on the other hand show signs of oxidation as they are heated above 700 ºC. In this case the elevated temperatures cause the opening of the end caps allowing the uptake of water. Oxygen exposure experiments indicate that degradation in field emission is two-fold and is ultimately dependent on the emission current at which the tip is operated. At low emission currents the degradation is exclusively due to oxidation. At high emission currents ion bombardment results in the degradation of the emitter. In between the two extremes, molybdenum tips are capable of stable emission.

This thesis deals with the topic of high resolution spectral analysis. In the first part, selected methods are presented and afterwards compared based on the Matlab implementations. The problematics of reduction of crossterms in quadratic time–frequency distributions is also covered. The second part is focused on the implementation and optimization of the algorithm for real-time computation of smoothed Wigner distribution function.

Abstract The Swedish railway has during many years received low customer satisfaction scores. Reasons for this are reliability, service and information concerns. The focus of this master thesis lies on the railway platform. The platform is an important part to the factors above as its information and facilities, which constitute a part of the service before the trip, could contribute in causing delays and reliability issues. This study investigates the behaviour of passengers on railway platforms and the effect that information, seats and entrances along the platform has on that behaviour. The goal is to find advantages and disadvantages and propose improvements in a Swedish setting. Platforms and trains that have been studied are in Stockholm, Sweden and Tokyo, Japan. Platform distributions and individual behaviour on the platforms are recorded with three different and relatively easy methods. Results of the Swedish study show aggregate trends where passengers tend to cluster around entrances and seats. Many Swedish passengers do not believe themselves to have enough information to know where their car will stop. These passengers do mostly not know that information exist and on average stand much further away from their closest door, when the train arrives, compared to passengers that believe themselves to know where their car will stop along the platform. The results of the Japanese study show that almost all passengers stand close to their door. From the findings in not only this, but also other studies, four aggregate properties in order to get better railway platforms are finally presented. These properties are reliability, clarity and availability of information and good location of seats and other facilities. Suggestions for tangible improvements for the Swedish railway platforms are finally given. Among these improvements are new information screens and standardized information and stopping position of trains.

For many decades the genus Milnesium was thought to consist of a single, cosmopolitan species: Milnesium tardigradum Doyere, 1840. However, recently the genus has been re-evaluated, and numerous new species have been described. Cur-rently, over twenty extant species and one fossil are recognised, and most appear to have very narrow geographic ranges. It is doubtful that M. tardigradum sensu stricto is truly cosmopolitan, but to evaluate this hypothesis, specimens previously identified as M. tardigradum must be re-examined using newly proposed taxonomic characters. As part of the All Taxa Biodiversity Inventory (ATBI) we collected Milnesium specimens from various locations in the Great Smoky Mountains National Park (GSMNP). Two Milnesium species have been evaluated, and one of them, Milnesium bohleberi sp. nov., is new to science. The new species is most similar to M. eurystomum but differs by shorter claws and a shorter, narrower, and more cylindrical buccal tube. The other Milnesium species, very rare in our collection, is morphologically indistin-guishable from Milnesium granulatum Ramazzotti 1962, which was previously known only from Chile, Italy and Roma-nia. Based on the recently revised description of M. tardigradum sensu stricto, this nominal species for the genus has not been found in the GSMNP samples.

<p style="text-align: justify;">In this doctoral dissertation several integral transforms are discussed.The first one is the Short time Fourier transform (STFT). We present continuity theorems for the STFT and its adjoint on the test function space <em>K</em>₁(ℝⁿ) and the topological tensor product <em>K</em>₁(ℝⁿ) &otimes; <em>U</em>(<strong>ℂ</strong>ⁿ), where <em>U</em>(<strong>ℂ</strong>ⁿ) is the space of entirerapidly decreasing functions in any horizontal band of&nbsp;<strong>ℂ</strong>ⁿ. We then use such continuity results to develop a framework for the STFT on K&#39;₁(ℝⁿ). Also, we devote one section to the characterization of <em>K</em>&rsquo;₁(ℝⁿ) and related spaces via modulation spaces. We also obtain various Tauberian theorems for the short-time Fourier transform.</p><p style="text-align: justify;">Part of the thesis is dedicated to the ridgelet and the Radon transform. We define and study the ridgelet transform of (Lizorkin) distributions and we show that the ridgelet transform and the ridgelet synthesis operator can be extended as continuous mappings <em>R</em>_{<em>&psi;&nbsp;</em>}: <em>S</em>&rsquo;₀(ℝⁿ) &rarr; <em>S</em>&rsquo;(<strong>Y</strong>ⁿ⁺¹) and <em>R^t</em>_{<span style="vertical-align: sub;">&psi;</span>}: <em>S</em>&rsquo;(<strong>Y</strong>ⁿ⁺¹) &rarr; <em>S</em>&rsquo;₀(ℝⁿ). We then use our results to develop a distributional framework for the ridgelet transform that is, we treat the ridgelet transform on <em>S</em>&rsquo;₀(ℝⁿ) via a duality approach. Then, the continuity theorems for the ridgelet transform are applied to discuss the continuity of the Radon transform on these spaces and their duals. Finally, we deal with some Abelian and Tauberian theorems relating the quasiasymptotic behavior of distributions with the quasiasymptotics of the its Radon and ridgelet transform.</p><p style="text-align: justify;">The last chapter is dedicated to the MRA of M-exponential distributions. We study the convergence of multiresolution expansions in various test function and distribution spaces and we discuss the pointwise convergence of multiresolution expansions to the distributional point values of a distribution. We also provide a characterization of the quasiasymptotic behavior in terms of multiresolution expansions and give an MRA sufficient condition for the existence of &alpha;-density points of positive measures.</p> / <p>U ovoj doktorskoj disertaciji razmotreno je nekoliko integralnih transformacija. Prva je short time Fourier transform (STFT). Date su i dokazane teoreme o neprekidnosti STFT i njena sinteza na prostoru test funkcije <em>K</em>₁(ℝⁿ) i na prostoru <em>K</em>₁(ℝⁿ) &otimes; <em>U</em>(ℂⁿ), gde je&nbsp;<em>U</em>(ℂⁿ) prostor od celih brzo opadajućih funkcija u proizvoljnom horizontalnom opsegu na ℂⁿ. Onda, ovi rezultati neprekidnosti su iskori&scaron;teni za razvijanje teorije STFT na prostoru <em>K</em>&rsquo;₁(ℝⁿ). Jedno poglavlje je posvećeno karakterizaciji&nbsp;<em>K</em>&rsquo;₁(ℝⁿ) sa srodnih modulaciskih prostora. Dokazani su i različiti Tauberovi rezultata za STFT. Deo teze je posvećen na ridglet i Radon transformacije. Ridgelet transformacija je definisana na (Lizorkin) distribucije i pokazano je da ridgelet transformacija i njen operator sinteze mogu da se pro&scaron;ire kako neprekidna preslikava <em>R</em>_&psi; : <em>S</em>&rsquo;₀(ℝⁿ) &rarr; <em>S</em>&rsquo;(<strong>Y</strong>ⁿ⁺¹) and <em>R</em>^t_&Psi;: <em>S</em>&rsquo;(<strong>Y</strong>ⁿ⁺¹) &rarr; <em>S</em>&rsquo;₀(ℝⁿ).&nbsp;Ridgelet transformacija na <em>S</em>&rsquo;₀(ℝⁿ) je data preko dualnog pristupa. Na&scaron;e teoreme neprekidnosti ridgelet transformacije su primenjene u dokazivanju neprekidnosti Radonove transformacije na Lizorkin test prostorima i njihovim dualima. Na kraju, dajemo Abelovih i Tauberovih teorema koji daju veze izmedju kvaziasimptotike distribucija i kvaziasimptotike rigdelet i Radonovog transfomaciju.<br />Zadnje poglavje je posveceno multirezolucijskog analizu M - eksponencijalnih distrubucije. Proucavamo konvergenciju multirezolucijkog razvoja u razlicitih prostori test funkcije i distribucije i razmotrena je tackasta konvergencija multirezolucijkog razvoju u tacku u distributivnog smislu. Obezbedjena je i karakterizacija kvaziasimptotike u pogled multirezolucijskog razvoju i dat dovoljni uslov za postojanje &alpha;-tacka gustine za pozitivne mere.</p>

<p>H-mere i H-distribucije su mikrolokalni objekti koji se koriste za ispitivanje jake konvergencije slabo konvergentnog niza u prostorima Lebega i prostorima Soboljeva. H-mere su uveli Tartar i&nbsp; Zerar (koji ih zove mikrolokalne mere defekta), u radovima [34] i [19]. H-mere su Radonove mere koje daju informacije o mogu &acute; cim oblastima jake konvergencije slabo konvergentnog<em> L</em>² niza. Da bismo mogli da posmatramo i slabo konvergentne<em> L</em>^p nizove za 1 &lt; p &lt; &infin;, Antonić&nbsp; i Mitrović u radu [11] uvode H-distribucije.</p><p>U disertaciji dajemo konstrukciju H-distribucija za slabo konvergentne nizove u <em>W</em>^-k,p prostorima, kad je 1 &lt; p &lt; &infin;, k &isin; ℕ&nbsp;i pokazujemo da kada je H-distribucija pridružena slabo konvergetnim nizovima jednaka nuli za sve test funkcije, onda imamo lokalno jaku konverenciju datog niza.</p><p>Takođe je pokazan i lokalizacijski princip, koji nam daje oblast u kojoj imamo lokalno jaku&nbsp; konvergenciju slabo konvergentnog niza. H-mere i H-distribucije deluju na test funkcije &phi;&nbsp;i &psi;&nbsp;(odgovarajuće regularnosti) koje su definisane na ℝ^d i S^d-1 (jedinična sfera u ℝ^d), pri&nbsp; čemu je funkcija &psi;, koju zovemo množilac, ograničena. U disertaciji uvodimo i H-distribucije sa neograničenim simbolom, pri čemu posmatramo slabo&nbsp; konvergentne nizove u Beselovim H^p_-s prostorima, gde je 1 &lt; p &lt; &infin;; s &isin; ℝ. U ovom delu koristimo teoriju pseudo-diferencijalnih operatora i dokazujemo kompaktnost komutatora [<i>A</i>_&psi;, T_&phi;] za razne klase množioca &psi;,&nbsp; &scaron;to je potrebno za dokaz postojanja H-distribucija. Takođe pokazujemo odgovarajuću verziju lokalizacijskog principa.</p> / <p>H-measures and H-distributions are microlocal tools that can be used to investigate strong conver-gence of weakly convergent sequences in the Lebesgue and Sobolev spaces.</p><p>H-measures are introduced by Tartar and G&eacute;rard (as microlocal defect measures) in papers [34] and [19]. H-measures are Radon measures and they provide information about the set of points where given weakly convergent sequence in <em>L</em>² converges strongly. In paper [11], Antonić and Mitrović introduced&nbsp; H-distributions in order to work with weakly convergent <em>L</em>^p sequences.</p><p>In this thesis we give construction of H-distributions for weakly convergent <em>W^-</em>^k,p sequences, where 1 &lt; p &lt; &infin;; k &isin;&nbsp;N. We show that if the H-distribution corresponding to given weakly convergent sequence is equal to zero, then we have locally strong convergence of the sequence. We also prove localization principle.</p><p>H-measures and H-distributions act on test functions &phi; and &psi;&nbsp;(regular enough) which are defined on ℝ^d and ^d-1 (unit sphere in ℝ^d ) and the function &psi;, which is called multiplier, is bounded. We also introduce H-distributions with unboundedmultipliers and in this&nbsp; case we assume that weakly convergent sequences are in Bessel potential spaces H^p_-s , where 1 &lt; p &lt; &infin;, s &isin; ℝ. Theory of pseudo-differential operators is used in construction of H-distributions with unbounded multipliers. We prove compactness of the commutator [<em>A</em>_{<em>&psi;</em>},T_&phi; ] for different classes of multipliers y and appropriate version of localization principle.</p>

La diffusion Compton virtuelle (DVCS) est un processus idéal pour étudier la structure interne du proton. Cette réaction exclusive permet d’accéder aux distributions de partons généralisées (GPDs) qui encodent les corrélations entre impulsion longitudinale et position transverse des partons à l’intérieur du proton. Le DVCS consiste à sonder le proton au moyen d’un photon virtuel de grande virtualité pour produire dans l’état final un unique photon réel de grande énergie tout en laissant le proton intact.A COMPASS au CERN, où deux années de données ont été collectées en 2016 et 2017 afin de mesurer la section efficace du processus DVCS, le photon virtuel est issu de la diffusion d’un faisceau de μ⁺ ou de μ⁻ polarisé de 160 GeV sur une cible d’hydrogène liquide. Toutes les particules de la réaction sont détectées dans l’expérience : le muon incident est détecté dans le télescope du faisceau, le muon diffracté et le photon réel sont détectés à l’avant dans le spectromètre et les trois calorimètres tandis que le proton de recul est détecté dans un détecteur de temps de vol placé autour de la cible.Je présente dans cette thèse l’état de l’analyse du processus DVCS sur les données collectées à COMPASS en 2016. Après un rappel du contexte théorique et expérimental, je décris l’expérience COMPASS. Je détaille ensuite mon travail de calibration du détecteur de proton de recul et de détermination de la position exacte de la cible de 2 cm de diamètre et 2.5 m de longueur. J’étudie dans la partie suivante la sélection de différents canaux de physique permettant de contrôler de manière systématique la qualité des détecteurs : la diffusion profondément inélastique (DIS) qui implique le télescope du faisceau et le spectromètre, la production exclusive de ρ⁰ qui inclut aussi le détecteur de temps de vol ; puis je présente la première analyse de la production exclusive de photons uniques qui implique en plus les trois calorimètres. Dans une dernière partie j’évoque les étapes nécessaires à la détermination de la section efficace du DVCS à partir de cette sélection, et je présente les premiers résultats issus de la simulation associée. / Virtual Compton Scattering (DVCS) is an ideal process to study the internal structure of proton. This exclusive reaction provides access to generalised parton distributions (GPDs), which encode the correlations between longitudinal momentum and transverse position of partons inside the proton. DVCS consists in probing a proton with a virtual photon of high virtuality, in order to produce a single high energy real photon while leaving the proton intact in the final state.At COMPASS at CERN, where two years of data were collected in 2016 and 2017 to measure the DVCS cross section, the virtual photon is produced by scattering of a 160 GeV polarised μ⁺ or μ⁻ beam on a liquid hydrogen target. All particles are detected in the experiment: the incident muon is detected in the beam telescope, the diffracted muon and the real photon are detected in the forward spectrometer and the three calorimeters, while the recoil proton is detected in a time-of-flight detector positioned around the target.In this thesis I present the state of the analysis of the DVCS process on the data collected at COMPASS in 2016. After a reminder of the theoretical and experimental context, I describe the COMPASS experiment. I then detail my work on calibrating the recoil proton detector and determining the the exact position of the 2 cm diameter and 2.5 m long target. In the next section, I study the selection of different physics channels used to systematically control detector quality: Deep Inelastic Scattering (DIS) which involves the beam-telescope and spectrometer, exclusive ρ⁰ production which requires the addition of the time-of-flight detector and I follow with the first analysis of the exclusive single photon production which depends as well on the calorimetres quality. In a last part, I discuss the necessary steps needed to extract the DVCS cross-section out of this event selection, and present the first results associated to the Monte-Carlo simulation.

Profile analysis is a multivariate statistical method for comparing the mean vectors for different groups. It consists of three tests, they are the tests for parallelism, level and flatness. The results from each test give information about the behaviour of the groups and the variables in the groups. The test statistics used when there are more than two groups are likelihood-ratio tests. However, issues in the form indeterminate test statistics occur in the high-dimensional setting, that is when there are more variables than observations. This thesis investigates a method to approach this problem by reducing the dimensionality of the data using scores, that is linear combinations of the variables. Three different ways of choosing this score are compared: the eigendecomposition and two variations of the non-negative matrix factorization. The methods are compared using simulations for five different type of mean parameter settings. The results show that the eigendecomposition is the best technique for choosing the score, and that using more scores only slightly improves the results. Moreover, the results for the parallelism and the flatness tests are shown to be very good, but the results for the level hypothesis deviate from the expectation.