Diploma thesis abstract Thesis title: Statistical inference in multivariate distributions based on copula models Author: Vojtěch Kika This diploma thesis aims for statistical inference in copula based models. Ba- sics of copula theory are described, followed by methods for statistical inference. These are divided into three main groups. First of them are parametric methods for copula parameter estimation which assume fully parametric structure, thus for both joint and marginal distributions. The second group consists of semi- parametric methods for copula parameter estimation which, unlike parametric methods, do not require parametric structure for marginal distributions. The last group describes goodness-of-fit tests used for testing the hypothesis that consi- dered copula belongs to some specific copula family. The thesis is accompanied by a simulation study that investigates the dependence of the observed coverage of the asymptotic confidence intervals for copula parameter on the sample size. Pseudolikelihood method was chosen for the simulation study since it is one of the most popular semiparametric methods. It is shown that sample size of 50 seems to be sufficient for the observed coverage to be close to the theoretical one. For Frank and Gumbel-Hougaard copula families even sample size of 30 gives us...

Une classe générale de statistiques de rangs basées sur la fonction caractéristique est introduite afin de tester l'hypothèse composite d'appartenance à une famille de copules multidimensionnelles. Ces statistiques d'adéquation sont définies comme des distances fonctionnelles de type L_2 pondérées entre une version non paramétrique et une version semi-paramétrique de la fonction caractéristique que l'on peut associer à une copule. Il est démontré que ces statistiques de test se comportent asymptotiquement comme des V-statistiques dégénérées d'ordre quatre et que leurs lois limites s'expriment en termes de sommes pondérées de variables khi-deux indépendantes. La convergence des tests sous des alternatives générales est établie, de même que la validité du bootstrap paramétrique pour le calcul de valeurs critiques. Le comportement des nouveaux tests sous des tailles d'échantillons faibles et modérées est étudié à l'aide de simulations et est comparé à celui d'un test concurrent fondé sur la copule empirique. La méthodologie est finalement illustrée sur un jeu de données à plusieurs dimensions.

L’émergence des technologies Big Data et Cloud computing pour répondre à l’accroissement constant de la complexité et de la diversité des données constituent un nouvel enjeu de taille pour les entreprises qui, désormais, doivent prendre en compte ce nouveau paradigme. Les opérateurs de services mobiles sont un exemple de sociétés qui cherchent à valoriser et monétiser les données collectées de leur utilisateurs. Cette recherche a pour objectif d’analyser ce nouvel enjeu qui allie d’une part l’explosion du nombre des données à analyser, et d’autre part, la constante émergence de nouvelles technologies et de leur adoption. Dans cette thèse, nous abordons la question de recherche suivante: « Dans quelle mesure les technologies Cloud Computing et Big Data contribuent aux tâches menées par les Data Scientists? » Sur la base d’une approche hypothético-déductive relayée par les théories classiques de l’adoption, les hypothèses et le modèle conceptuel sont inspirés du modèle de l’adéquation de la tâche et de la technologie (TTF) de Goodhue. Les facteurs proposés incluent le Big Data et le Cloud Computing, la tâche, la technologie, l'individu, le TTF, l’utilisation et les impacts réalisés. Cette thèse aborde sept hypothèses qui adressent spécifiquement les faiblesses des modèles précédents. Une enquête a été conduite auprès de 169 chercheurs contribuant à l’analyse des données mobiles. Une analyse quantitative a été effectuée afin de démontrer la validité des mesures effectuées et d’établir la pertinence du modèle théorique proposé. L’analyse partielle des moindres carrés a été utilisée (partial least square) pour établir les corrélations entre les construits. Cette recherche délivre deux contributions majeures : le développement d'un construit (TTF) spécifique aux technologies Big Data et Cloud computing ainsi que la validation de ce construit dans le modèle d’adéquation des technologies Big data - Cloud Computing et de l’analyse des données mobiles. / A new economic paradigm is emerging as a result of enterprises generating and managing increasing amounts of data and looking for technologies like cloud computing and Big Data to improve data-driven decision making and ultimately performance. Mobile service providers are an example of firms that are looking to monetize the collected mobile data. Our thesis explores cloud computing determinants of adoption and Big Data determinants of adoption at the user level. In this thesis, we employ a quantitative research methodology and operationalized using a cross-sectional survey so temporal consistency could be maintained for all the variables. The TTF model was supported by results analyzed using partial least square (PLS) structural equation modeling (SEM), which reflects positive relationships between individual, technology and task factors on TTF for mobile data analysis.Our research makes two contributions: the development of a new TTF construct – task-Big Data/cloud computing technology fit model – and the testing of that construct in a model overcoming the rigidness of the original TTF model by effectively addressing technology through five subconstructs related to technology platform (Big Data) and technology infrastructure (cloud computing intention to use). These findings provide direction to mobile service providers for the implementation of cloud-based Big Data tools in order to enable data-driven decision-making and monetize the output from mobile data traffic analysis.

Master of Science / Department of Statistics / James W. Neill / Cluster based lack of fit tests for linear regression models are generally effective in detecting model inadequacy due to between- or within-cluster lack of fit. Typically, lack of fit exists as a combination of these two pure types, and can be extremely difficult to detect depending on the nature of the mixture. Su and Yang (2006) and Miller and Neill (2007) have proposed lack of fit tests which address this problem. Based on a simulation comparison of the two testing procedures, it is concluded that the Su and Yang test can be expected to be effective when the true model is locally well approximated within each specified cluster and the lack of fit is not due to an unspecified predictor variable. The Miller and Neill test accommodates a broader alternative, which can thus result in comparatively lower but effective power. However, the latter test demonstrated the ability to detect model inadequacy when the lack of fit was a function of an unspecified predictor variable and does not require a specified clustering for implementation.

Master of Science / Department of Statistics / Shie-Shien Yang / In fitting linear models, R[superscript]2 statistic has been wildly used as one of the measures to assess the goodness-of-fit and prediction power of the model. Unlike fixed linear models, at this time there is no single universally accepted measure for assessing goodness-of-fit and prediction power of a linear mixed model. In this report, we reviewed seven different approaches proposed to define a measure analogous to the usual R[superscript]2 statistic for assessing mixed models. One of seven statistics,Rc, has both conditional and marginal versions. Association mapping is an efficient way to link the genotype data with the phenotype diversity. When applying the R[superscript]2 statistic to the association mapping application, it can determine how well genetic polymorphisms, which are the explanatory variables in the mixed models, explain the phenotypic variation, which is the dependent variation. A linear mixed model method recently has been developed to control the spurious associations due to population structure and relative kinship among individuals of an association mapping. We assess seven definitions of R[superscript]2 statistic for the linear mixed model using data from two empirical association mapping samples: a sample with 277 diverse maize inbred lines and a global sample of 95 Arabidopsis thaliana accessions using the new method. R[superscript]2[subscript]LR statistic derived from the log-likelihood principle follows all the criterions of R[superscript]2 statistic and can be used to understand the overlap between population structure and relative kinship in controlling for sample relatedness. From our results,R[superscript]2[subscript]LR statistic is an appropriate R[superscript]2 statistic for comparing models with different fixed and random variables. Therefore, we recommend using RLR statistic for linear mixed models in association mapping.

Denna studie är en kandidatuppsats i designteknik som utförs på uppdrag av ett svenskt modeföretag för att utveckla nya livgrunder till oversize-plagg. Företaget har problem med överdelar på damavdelningen som lutar bakåt vilket visar sig främst på plagg med mycket rörelsevidd. Studien undersöker olika grunder ur litteratur som jämförs med två grunder från företaget, en med insnitt och en utan. Dessa modifieras sedan för att uppnå kriterierna för en god balans. Arbetet utförs genom mönsterkonstruktion i ett CAD-program, drapering och uppsömnad av toiler. Alla toiler provas av på tre till fem provmodeller i storlek 38 som passar företagets målgrupp samt på provdocka och virtuellt på en avatar i 3D. Skillnaderna konstruktionsmässigt analyseras i ett ritprogram där 2D mönstren placeras ovanpå varandra och jämförs. De största skillnaderna mellan litteraturens- och företagets grunder är axelsömmens lutning, axelsömmens placering och halsdjupet vilka justeras på företagets grunder. Form på axelsömmen visar sig i denna studie ha betydelse för att få till en bra balans samtidigt som halsen ligger slätt mot kroppen, främst på grunden utan insnitt. / This is a bachelor essay in garment technology which is performed on behalf of a Swedish fashion company to develop new top basic patterns for oversize garments.  The company is experiencing a problem among tops at the department for women with garments leaning backwards, specifically on garment with a lot of ease. This study examines different basic patters from literature and compares it with two basic patterns from the company, one with a dart and one without. These basic patters are later modified to fit the criteria of a good balance. The method chosen for this study is patternmaking in a CAD-program, draping and sewing of sample garments. All sample garments are fitted on three to five models in size 38 which fits the target group of the company. It is also tried on a dummy and virtually on a 3D avatar. Differences in the pattern are analyzed in a drawing program where the 2D pieces are placed on top of each other. The biggest difference between the basic pattern from the literature and the company is the slant of the shoulder seam, the placement of it and the depth of the neckline which is modified on the basic patters from the company. Shape on the shoulder seam shows significant meaning in achieving good balance as well as a neckline that lies flat against the body, especially on the basic pattern without a dart.

In many applications, e.g. medical, industrial and military systems, it is of interest to fit a circle to scattered data points belonging to a complete or incomplete circular arc. As an example, circle fitting can be applied in the industry, for quality control, when investigation is required to verify if a manufactured circular object has the desired radius or not. A variety of methods have been developed to handle the circle fitting problem. Some methods are relatively complex and provide more accurate circle fitting, whereas some are simple and fast but lacks accuracy. Furthermore, some methods handle circle fitting better on incomplete circular objects. However, for practical machine vision implementations, there seems to be a lack of study when it comes to circle fitting on largely incomplete circular arcs. Largely incomplete circular arcs refer to short arcs having corresponding angles of few degrees, e.g. less than 10°. Hence, this thesis deals with design and implementation of circle fitting on largely incomplete circular objects. The goal is to investigate the shortest circular arc, i.e. the shortest possible angle that, can be fitted to a circle with an accuracy of at least 98%. The approach includes studying related work, developing a vision based algorithm for circle fitting on incomplete circular objects and conducting experiments using live stream 2D images. We designed and implemented an algorithm, based on a circle fitting algorithm, called Hyper fit. Our experimental set-up, with a 5-Megapixel camera, showed that it is possible to fit a circle, with an accuracy of 98%, to a short circular arc with an angle of only 1.95° of a complete circle. 1.95° corresponds to 0.54% of a complete circles circumference. Results showed that, using a high-resolution camera, it is possible to fit accurate circles on largely incomplete circular arcs. Moreover, the implementation achieved the real-time requirement, as it could process at least 3 fps (frames per second).

[pt] As organizações sem fins lucrativos na última década cresceram em relevância no cenário mundial. Sua transformação em organizações estruturadas, de cujos resultados depende sua sustentabilidade, elevou a importância da sua gestão estratégica, com a incorporação de ferramentas gerenciais. A presente pesquisa buscou, por meio de uma perspectiva integrativa de gestão estratégica, identificar se há valor agregado do fit entre os fatores organizacionais específicos de três organizações sem fins lucrativos e suas ferramentas gerenciais de apoio ao planejamento organizacional. Foram estudados os planejamentos feitos entre 2004 e 2008 pelo CDI Matriz, com atuação na América Latina; pela Associação Saúde Criança Renascer, presente no Brasil, e pelo CDI Chile. Dados foram coletados por meio de investigação documental/telematizada e percepções por meio de questionário estruturado e entrevistas. Em seguida, dados e percepções foram analisados com base no referencial teórico investigado, especificamente pelo modelo genérico integrativo de Macedo-Soares. A pesquisa identificou como características organizacionais relevantes a falta de objetivos organizacionais claros, a falta de alinhamento entre as áreas e a preocupação com sustentabilidade direcionadora das organizações. Os resultados sugerem que as ferramentas promotoras da interdependência da gestão, da troca de informações e da divergência informada na tomada de decisão contribuíram para alinhar as áreas em torno de uma visão compartilhada, além de criar uma cultura de planejamento. / [en] Non-profit organizations have increased in worldwide relevance during the last decade. Their transformation in structured organizations, whose results impact their sustainability, increased the importance of strategic management, with the incorporation of managerial tools. This research sought, through an integrative perspective of strategic management, to identify if there is value of a fit between the specific organizational factors of three nonprofit organizations and their management tools to support organizational planning. The plans studied took place between 2004 and 2008 at CDI Matriz, with activities in Latin America, at Associação Saúde Criança Renascer, present in Brazil and at CDI Chile. Data was collected through documental/electronic research and perceptions through a structured questionnaire and interviews. Afterwards, data and perceptions were analyzed in reference to the theoretical framework investigated, specifically the Macedo-Soares generic integrative model. The research identified as relevant organizational characteristics the lack of clear organizational objectives, the lack of alignment between areas and the concern with sustainability directing the organizations. The results suggest that the tools that promote the interdependence of management, information exchange and the informed dissent in decision making contributed to align the areas around a shared vision, besides creating a planning culture.

This thesis is entitled The Role of Management in the Turnaround Process and is presented for the degree of Doctor of Business Administration at the University of Manchester by Ian Roberts in April 2015. Inquiry into turnaround has largely has been carried forward in two broad areas of study: strategy and management. Strategy research has attempted to identify a single or limited number of strategies which apply in all turnaround situations, while management studies have concentrated on managers themselves or some of the techniques they employ. Neither school has systematically analysed how turnarounds are actually implemented in dysfunctional organisations and both schools are weak in extant theory. As a consequence, a holistic explanation of the process leading from crisis to viability is lacking. This thesis addresses these lacunae in three ways. Firstly, it provides a systematic study of how professional turnaround managers implement the turnaround process over time. Secondly, it applies a unique and complementary mix of extant theory which addresses both cognitive and organisational aspects of strategy formulation and implementation. Thirdly, it presents a holistic turnaround model based on six core constructs which are argued to be necessary and sufficient to explain the dynamics of the turnaround process. The thesis adopts a mixed-methods approach. A survey is used in order to gather data from turnaround professionals on implementation methods. This data is then subjected to statistical analysis in order to identify the most important factors for implementation. These concepts are then blended into a conceptual framework which is tested for its explanatory and predictive power on a matched pair of turnaround case studies of two mid-sized UK manufacturing companies in the household goods sector, one of which executed a successful turnaround and one of which failed in the attempt. The model is able to explain and predict the outcome in both cases. The thesis is one of only a handful of turnaround studies which employs a rigorous case study research protocol. Its principal contribution is that of a middle range turnaround theory of the causal factors leading from organisational crisis to dissolution or viability.

Master of Science / Department of Computing and Information Sciences / Mitchell L. Neilsen / It is well recognized that staying fit and healthy is a significant public problem now days and engaging in moderate levels of physical activity and having healthy food is associated with positive outcomes. Many dietary applications are being used at present times, but they don’t include physical activity in their assessment and even if they do so they require user to manually enter the data which gives room for unintentional mistakes. Nutritional wizard is a dietary and nutritional coach that works synchronously with Fitbit data to achieve this. This application analyses the food intake of an individual and automatically inspects this with his physical activity recorded through the Fitbit to give a cover on his health and fitness. The application has four main parts, input food, analysis, profile, and recipes. In input food part, user gives all the details of the food he had on that day and nutritional wizard gives the person’s calorie intake, steps he made on that day, calories spent, excess calories stored, steps he has to still make on that day. Analysis part of the application gives graphical representations of daily calorie intake in a specific time frame given by user. Profile has user’s height and weight based on which BMI of user is calculated and is represented in a Gantt chart showing if he is in low, good or high BMI. Last Recipes module has few recipes categorized into different selections low carb, low fat, and low protein which user can use to decrease his calorie intake. The completely developed app can be used by any user maintaining a fit bit account to provide good nutritional benefits.