Quando considerado o funcionamento de uma aeronave em operação de voo para transporte de passageiros civis, há a ocorrência de falhas que podem afetar a aeronavegabilidade ou não e, dentro das falhas que não afetam,encontram-se,por exemplo,as falhas de recursos utilizados durante o voo pelos passageiros, como poltronas ou sistema de entretenimento de bordo. É notória a importância de uma análise em relação ao diagnóstico de falhas que possuem maior impacto na satisfação do cliente, assim como de uma tomada de medidas de manutenção preventiva para que tais ocorrências sejam minimizadas. Como resultado, entre as 16 principais falhas que ocorrem, apenas 4 falhas caracterizam-se como sendo de maior impacto. É nesse sentido que o principal objetivo deste trabalho é descrever o tempo de falha desses principais componentes. Para tanto, ajustamos diferentes distribuições de probabilidade que descrevem o comportamento dessas falhas.Levando-se em conta que em um tipo de falha os principais modelos utilizados na literatura como Weibull, Gamae Log-normal não são os mais adequados para descrever tal característica, propomos um novo modelo de confiabilidade intitulado Inversa Lindley Ponderada. As propriedades matemáticas e os métodos de estimação dos parâmetros dessa nova distribuição são exploradas ao longo dessa dissertação. O modelo é utilizado para descrever o tempo de vida de um tipo de falha, retornando melhores ajustes que as principais distribuições de confiabilidade. Com base nesses novos resultados, propomos uma política de manutenção preventiva para os principais componentes. / During the operation of anaircraft for comercial flight of passengers, there are occurrences of failures that can compromise the satisfaction of passengers. For instance, the malfunction of seats or the entertainment system. This problem is very important for the operation in order to maintain customer satisfaction and customer loyalty. Indeed, it is ofmain interesting analysis for diagnosis of failures, which have greater impact on customer satisfaction. The results of such analysis can be used to propose preventive maintenance schemes in order to minimize costs. Based on our analysis, among the 16 major types of failures that occur, 4 types showed to have greatest impacts in the role system. In this sense,the main objective of this work is to describe the failure time of these main components. To achieve that, we set different probability distributions to describe the random behavior of these failures. For one type of failure we showed that there is space for improvement of its probability distribution as the Weibull,Gamma and Log-normal are not the most adequate to describe the random behaviour of such characteristic. Therefore, we propose a new reliability distribution, the so called inverse Weighted Lindley distribution. The mathematical properties and the parameter estimation of the parameters for the proposed distribution are discussed. The new distribution is used to describe the lifetime of one of types of failure. Based on these new results, we propose a preventive maintenance policy to maintain the quality of the main components.

A ressonância magnética (RM) tem sido amplamente utilizada no diagnóstico por imagem de alterações de glândulas salivares. No entanto, a presença de aspectos similares nas imagens com técnicas convencionais de RM dificulta a distinção do diagnóstico entre patologias inflamatórias e neoplásicas. O objetivo deste estudo foi comparar valores dos coeficientes de difusão aparentes (ADC - Apparent Diffusion Coefficient) de imagem ponderada em difusão (DWI - Difusion Weighted Imaging) com ressonância magnética, entre glândulas salivares normais, casos com sialoadenite e com adenoma pleomórfico das glândulas salivares maiores. Vinte e dois pacientes (totalizando 44 glândulas salivares maiores) diagnosticados com sialoadenite unilateral (em glândula parótida ou submandibular) ou adenoma pleomórfico (apenas em parótida) foram selecionados. Todas as glândulas contralaterais não afetadas também foram analisadas. Imagens de RM ponderadas em T1, T2 e DWI foram obtidas utilizando sequências de pulso spin-eco (SE) com um aparelho de ressonância magnética de 1.5 Tesla. Primeiramente, a performance diagnóstica (sensitividade, especificidade e acurácia) foi calculada para três observadores após analisarem imagens de RM e DWI, separadamente. Em seguida, os valores médios de ADC foram comparados entre os três grupos analisados (glândulas normais contralateral, sialoadenite e adenoma pleomórfico). O uso da DWI rendeu uma melhor performance diagnóstica em geral para todos os observadores. Além disso, casos de adenoma pleomórfico apresentaram os maiores valores de ADC do estudo. Dentro das limitações deste estudo, os resultados sugerem que DWI permite a diferenciação entre sialoadenite e adenoma pleomórfico. / Alterations of the salivary glands are usually detected by conventional magnetic resonance imaging (MRI) techniques. However, their imaging presentation may present similar aspects. The aim of this study was to compare apparent diffusion coefficient (ADC) values from diffusion-weighted MRI (DWI) among normal salivary glands, cases with sialadenitis and with pleomorphic adenoma of major salivary glands. Twenty-two patients (totaling 44 major salivary glands) diagnosed with either unilateral sialadenitis (on either parotid or submandibular gland) or parotid gland pleomorphic adenoma were selected. Contralateral non-affected glands (normal) were also analyzed. DWI images were achieved using a spin-echo (SE) pulse sequence with a 1.5T MRI device. Mean ADC values were compared among the three groups analyzed (contralateral normal glands, sialadenitis and pleomorphic adenoma). Furthermore, diagnostic performance of MRI and DWI were calculated for three observers. DWI also presented better diagnostic performance results. In addition, cases of pleomorphic adenoma presented the highest ADC values of the study. Within the limitations of this study, the present results suggest that DWI allows for differentiation between parotid sialadenitis and pleomorphic adenoma.

We improved two beamforming-based methodologies for seismological analysis. The first one is a new Three-Dimensional Phase-Weighted Relative Back Projection (3-D PWBP) method to improve the spatial resolution of Back Projection results. We exploit both phase and amplitude of the seismogram signal to enhance the distinction of correlated signals. Also, we implement a 3-D velocity model to provide more accurate travel times. We vindicate these refinements with several synthetic tests and an analysis of the 1997 Mw 7.2 Zirkuh (Iran) earthquake, which we show ruptured mainly unilaterally southwards at a rupture speed of ∼3.0 km/s along its ∼125 km- long, mostly single-stranded surface rupture. Then, we apply the new method to the more complex case of the 2016 Mw 7.8 Kaikoura (New Zealand) earthquake, which we demonstrate is divided into two major stages separated by a gap of ∼8 s and ∼30–40 km. The overall rupture speed is ∼1.7 km/s and the overall duration is ∼84 s, considerably shorter than some earlier estimates. We see no clear evidence for continuous failure of the subduction interface that underlies the known, surface-rupturing crustal faults, though we cannot rule out its involvement in the second major stage in the northern part of the rupture area. The late (∼80 s) peak in relative energy is likely a high-frequency stopping phase, and the rupture appears to terminate southwest of the offshore Needles fault. The second methodology is a novel workflow for earthquake detection and location, named Seismicity-Scanning based on Navigated Automatic Phase-picking (S-SNAP). By taking a cocktail approach that combines Source-Scanning, Kurtosis-based Phase-picking and the Maximum Intersection location technique into a single integrated workflow, this new method is capable of delineating complex spatiotemporal distributions of seismicity. It is automatic, efficiently providing earthquake locations with high comprehensiveness and accuracy. We apply S-SNAP to a dataset recorded by a dense local seismic array during a hydraulic fracturing operation to test this novel approach and to demonstrate the effectiveness of S-SNAP in comparison to existing methods. Overall, S-SNAP found nearly four times as many high-quality events as a template-matching based catalogue. All events in the previous catalogue are identi- fied with similar epicenter, depth and magnitude, while no false detections are found by visual inspection. / Graduate

Despite the rich knowledge surrounding risk factors for malaria, the spatial processes of malaria transmission and vector control interventions are underexplored. This thesis aims 1) to describe the spatial variation of risk factor effects on malaria infection, and 2) to determine the presence and range of any community effect from malaria vector control interventions. Data from a cluster-randomized control trial in Tanzania were analyzed to determine the geographically-weighted odds of malaria infection in children at trial baseline and post-intervention. The spatial range of intervention effects on malaria infection was estimated post-intervention using semivariance models. Spatial heterogeneities in malaria infection and each covariate under study were found. The median effective semivariance range of intervention effects was approximately 1200 meters, suggesting the presence of a community effect that may cause contamination between trial clusters. Trials should consider these spatial effects when examining interventions and ensure that clusters are adequately insulated from contamination.

Radiological imaging already has a key role in the detection and management of patients with metastatic colorectal cancer (mCRC). With the evolution of personalised medicine there is a need for non-invasive imaging biomarkers that can detect early tumour response to targeted therapies. Translation from bench to bedside requires a multicentre approach that follows an agreed development roadmap to ensure that the proposed biomarker is precise (reproducible/ repeatable) and accurate in its characterisation of a meaningful physiological, pathological or post treatment response. The following thesis (organized in the alternative format with experimental studies written as individual complete manuscripts) investigates methods to improve precision and accuracy of the Apparent Diffusion Coefficient (ADC), a proposed quantitative imaging biomarker with a potential role in characterisation of post treatment responses in mCRC. The first objective was to establish baseline multicentre reproducibility (n=20) for ADC. A change in ADC greater than 21.1% was required to determine a post treatment response. Using a statistical error model, the dominating factors that influenced reproducibility were motion artefact and tumour volume. In the second study these factors were addressed using a single centre cohort with pre and post treatment data. Correcting for errors due to motion and tumour volume improved sensitivity from 30.3% to 1.7%, so a post treatment response was detected in 6/12 tumours compared to 0/12 using the baseline approach. In the third study, motion correction was implemented and the statistical error model was applied successfully to a multicentre cohort of 15 patients (1.9% sensitivity). The results of this thesis highlights that with careful consideration and correction of factors that negatively influence sensitivity, ADC is a potential imaging biomarker for use in post treatment response for patients with mCRC.

Governance of the healthcare systems is one of the most important challenges forWestern countries. Within this, an accurate assessment of the quality is key to policy makers and public managers, in order to guarantee equity, effectiveness and efficiency. In this thesis, we investigate aspects and methods related to healthcare evaluation by focussing on the healthcare system in Lombardy (Italy), where public and private providers compete with each other, patients are free to choose where to be hospitalized, and a pay-for-performance program was recently implemented. The general aim of this thesis is to highlight the role of statistics within a quality evaluation framework, in the form of advancing the statistical methods used to measure quality, of evaluating the effectiveness of implemented policies, and of testing the effect that mechanisms of competition and cooperation can have on the quality of a healthcare system. We firstly advance a new methodological approach for measuring hospital quality, providing a new tool for managers involved in performance evaluations. Multilevel models are typically used in healthcare, in order to account for the hierarchical structure of the data. These models however do not account for unobserved heterogeneity. We therefore propose an extension of the cluster-weighted models to the multilevel framework and focus in particular on the case of a binary dependent variable, which is common in healthcare. The resulting multilevel logistic cluster-weighted model is shown to perform well in a healthcare evaluation context. Secondly, we evaluate the effectiveness of a pay-for-performance program. Differently from the existent literature, in this thesis we evaluate this program on the basis of five health outcomes and across a wide range of medical conditions. Availability of data pre and post-policy in Lombardy allows us to use a difference-in-differences approach. The statistical model includes multiple dependent outcomes, that allow quantifying the joint effect of the program, and random effects, that account for the heterogeneity of the data at the ward and hospital level. The results show that the policy has overall a positive effect on the hospitals' performance. Thirdly, we study the effect of pro-competition reforms on the hospital quality. In Lombardy, competition between hospitals has been mostly driven by the adoption of a quasi-market system. Our results show that no association exists between hospital quality and competition. We speculate that this may be the result of asymmetric information, i.e. the lack of transparent information provided to citizens about the quality of hospitals. This is bound to reduce the impact of pro-competition reforms on quality and can in part explain the conflicting results found in the literature on this subject. Our results should motivate a public disclosure of quality evaluations. Regardless of the specifics of a system, hospitals are altruistic economic agents and they cooperate in order to improve their quality. In this work, we analyse the effect of cooperation on quality, taking the network of patients' transfers between hospitals as a proxy of their level of cooperation. Using the latest network models, we find that cooperation does lead to an increase in quality and should therefore be encouraged by policy makers.

La vérification automatique est aujourd'hui devenue un domaine central de recherche en informatique. Depuis plus de 25 ans, une riche théorie a été développée menant à de nombreux outils, à la fois académiques et industriels, permettant la vérification de propriétés booléennes - celles qui peuvent être soit vraies soit fausses. Les besoins actuels évoluent vers une analyse plus fine, c'est-à-dire plus quantitative. L'extension des techniques de vérification aux domaines quantitatifs a débuté depuis 15 ans avec les systèmes probabilistes. Cependant, de nombreuses autres propriétés quantitatives existent, telles que la durée de vie d'un équipement, la consommation énergétique d'une application, la fiabilité d'un programme, ou le nombre de résultats d'une requête dans une base de données. Exprimer ces propriétés requiert de nouveaux langages de spécification, ainsi que des algorithmes vérifiant ces propriétés sur une structure donnée. Cette thèse a pour objectif l'étude de plusieurs formalismes permettant de spécifier de telles propriétés, qu'ils soient dénotationnels - expressions régulières, logiques monadiques ou logiques temporelles - ou davantage opérationnels, comme des automates pondérés, éventuellement étendus avec des jetons. Un premier objectif de ce manuscript est l'étude de résultats d'expressivité comparant ces formalismes. En particulier, on donne des traductions efficaces des formalismes dénotationnels vers celui opérationnel. Ces objets, ainsi que les résultats associés, sont présentés dans un cadre unifié de structures de graphes. Ils peuvent, entre autres, s'appliquer aux mots et arbres finis, aux mots emboîtés (nested words), aux images ou aux traces de Mazurkiewicz. Par conséquent, la vérification de propriétés quantitatives de traces de programmes (potentiellement récursifs, ou concurrents), les requêtes sur des documents XML (modélisant par exemple des bases de données), ou le traitement des langues naturelles sont des applications possibles. On s'intéresse ensuite aux questions algorithmiques que soulèvent naturellement ces résultats, tels que l'évaluation, la satisfaction et le model checking. En particulier, on étudie la décidabilité et la complexité de certains de ces problèmes, en fonction du semi-anneau sous-jacent et des structures considérées (mots, arbres...). Finalement, on considère des restrictions intéressantes des formalismes précédents. Certaines permettent d'étendre l'ensemble des semi-anneau sur lesquels on peut spécifier des propriétés quantitatives. Une autre est dédiée à l'étude du cas spécial de spécifications probabilistes : on étudie en particulier des fragments syntaxiques de nos formalismes génériques de spécification générant uniquement des comportements probabilistes. / Automatic verification has nowadays become a central domain of investigation in computer science. Over 25 years, a rich theory has been developed leading to numerous tools, both in academics and industry, allowing the verification of Boolean properties - those that can be either true or false. Current needs evolve to a finer analysis, a more quantitative one. Extension of verification techniques to quantitative domains has begun 15 years ago with probabilistic systems. However, many other quantitative properties are of interest, such as the lifespan of an equipment, energy consumption of an application, the reliability of a program, or the number of results matching a database query. Expressing these properties requires new specification languages, as well as algorithms checking these properties over a given structure. This thesis aims at investigating several formalisms, equipped with weights, able to specify such properties: denotational ones - like regular expressions, first-order logic with transitive closure, or temporal logics - or more operational ones, like navigating automata, possibly extended with pebbles. A first objective of this thesis is to study expressiveness results comparing these formalisms. In particular, we give efficient translations from denotational formalisms to the operational one. These objects, and the associated results, are presented in a unified framework of graph structures. This permits to handle finite words and trees, nested words, pictures or Mazurkiewicz traces, as special cases. Therefore, possible applications are the verification of quantitative properties of traces of programs (possibly recursive, or concurrent), querying of XML documents (modeling databases for example), or natural language processing. Second, we tackle some of the algorithmic questions that naturally arise in this context, like evaluation, satisfiability and model checking. In particular, we study some decidability and complexity results of these problems depending on the underlying semiring and the structures under consideration (words, trees...). Finally, we consider some interesting restrictions of the previous formalisms. Some permit to extend the class of semirings on which we may specify quantitative properties. Another is dedicated to the special case of probabilistic specifications: in particular, we study syntactic fragments of our generic specification formalisms generating only probabilistic behaviors.

Title: Methods of dynamical analysis of portfolio composition Author: Ivana Meňhartová Department: Department of Probability and Mathematical Statistics Supervisor: Mgr. Tomáš Hanzák, KPMS, MFF UK Abstract: In the presented thesis we study methods used for dynamic analysis of portfolio based on it's revenues. The thesis focuses on Kalman filter and local- ly weighted regression as two basic methods for dynamic analysis. It describes in detail theory for these methods as well as their utilization and it discusses their proper settings. Practical applications of both methods on artificial data and real data from Prague stock-exchange are presented. Using artificial data we demonstrate practical importance of Kalman filter's assumptions. Afterwards we introduce term multicolinearity as a possible complication to real data applicati- ons. At the end of the thesis we compare results and usage of both methods and we introduce possibility of enhancing Kalman filter by projection of estimations or by CUSUM tests (change detection tests). Keywords: Kalman filter, locally weighted regression, multicollinearity, CUSUM test

This study develops a panel data analysis over 27 transition and post transition economies for the period 2003-2010. Its intent is to investigate empirically the true effect of seven variables into foreign flows and takes later on the advantage of observed findings to conduct a comparative analysis between Kosovo and regional countries such: Albania, Bosnia and Herzegovina, Macedonia, Montenegro and Serbia. As the breakdown period (2008-2010) was included in the data set used to modelling the behaviour of FDI, both Chow test and the time dummies technique suggest the presence of structural break. Ultimately, empirical results show that FDI is positively related with one year lagged effect of real GDP growth, trade openness, labour force, low level of wages proxied by remittances, real interest rate and the low level of corruption. Besides, the corporate income tax is found to be significant and inversely related with foreign flows. The comparative analysis referring the growth rate of real GDP shows that Kosovo has the most stable macroeconomic environment in the region, but still it is continuously confronted by the high deficit of trade balance and high rate of unemployment. Appart, the key obstacle that has abolished efforts for foreign investment attraction is found to be the trade blockade of...

The logistical complexities of performing multi-site longitudinal diffusion-weighted imaging (DWI) studies requires careful construction of analysis tools and procedures. Proposed clinical trials for therapies in neurodegenerative disease are known to re- quire several hundred subjects, thus mandating multiple site participation to obtain sufficient sample sizes. DWI is an important tool for monitoring diffusivity properties of white matter (WM) in disease progression. The multi-site nature of clinical trials requires new strategies in DWI processing and analysis to reliably measure longitudi- nal WM changes. This work describes the process of developing and validating robust analysis methodologies to process multi-site DWI data in a rare, neurodegenerative disease. Key processing components to accomplish a robust DWI processing system include: DICOM conversion, automated quality control, unbiased atlas construction, fiber tracking, and statistical analysis. Extensive validation studies were performed to characterize methodological results within and across the common confounds inherent in multi-site clinical trials. The conversion and automated quality control tools optimized for this work both enhanced the ability to reliably obtain repeat diffusion tensor image (DTI) scalar measurements in a reliability analysis of healthy controls scanned at multiple sites using multiple scanner vendors. A DTI scalar analysis performed on focused WM regions showed it was possible to detect significant mean differences of DTI scalars among separate groups of a neurodegenerative disease population. The DTI scalar analysis paved the way for an atlas-based cross-sectional fiber tracking analysis. In the cross-sectional fiber tracking analysis, multi-site data was brought into the same space, making major fiber tracts terminating in the focused WM regions of the scalar analysis from all participants comparable. Significant differences in diffusivity were found throughout each tract among separate groups of the neurodegenerative disease population. In addition, multiple neuropsychological cognitive variables that have a documented ability to track disease progression of the neurodegenerative disease, strongly correlated with many of the DTI scalars in each tract. The findings of the cross-sectional fiber tracking analysis were reinforced by similar findings produced by a longitudinal fiber tracking analysis. Collectively, these findings suggest that cogni- tive deficits seen in the neurodegenerative disease population could be explained by changes in diffusivity of the tracts explored in this work. In addition to the longi- tudinal fiber tracking analysis examining diffusivity, methods for a WM morphology analysis using parallel transport to apply longitudinal volume changes to a template was explored.