We propose methods for analyzing longitudinal data, obtained in clinical trials and other applications with repeated measures of responses taken over time. Common characteristics of longitudinal studies are correlated responses and observations taken at unequal points in time. The first part of this dissertation examines the justification of a block bootstrap procedure for the repeated measurement designs, which takes into account the dependence structure of the data by resampling blocks of adjacent observations rather than individual data points. In the case of dependent stationary data, under regular conditions, the approximately studentized or standardized block bootstrap possesses a higher order of accuracy. With longitudinal data, the second part of this dissertation shows that the diagonal optimal weights for unbalanced designs can be made to improve the efficiency of the estimators in terms of mean squared error criterion. Simulation study is conducted for each of the longitudinal designs. We will also analyze repeated measurement data set concerning nursing home residents with multiple sclerosis, which is obtained from a large database termed the minimum data set (MDS).

Social networks are the current phenomenon, and their integration begins to gain importance. Basic idea is to pair identical pieces of information stored in various social networks and to detect inconsistencies between them. The most important data that need to be integrated are user profi les and lists of contacts. In our work, we propose the integration of both the pro files and lists of contacts, and even groups of users that could be created in social networks. Implementation is not part of the work. However the proposal suggests the creation of Main social network in form of a web application, that brokers the integration to end users. From the perspective of users it is bene cial to display data from all social networks in one place and automate detection of di erences. User's data could be used for commercial purposes, of course, within legal limits. From this perspective, the biggest contribution of Main social network is a social graph. It consists of various social graphs of social networks. Additionally it would contain information about membership of users in social networks. The proposal consists of three main parts: main processes of the Main social network, its data model and means of gathering information from different networks. The design of data model was partially inspired by existing...

Increasingly, the health care field is utilizing information technology (IT) to help manage large volumes of medical data. This has created a need for health care workers to learn IT skills, which include information gathering skills (IGS), information analysis skills (IAS), and technology skills (TS). Research focused on medical students learning IT skills seems limited, while research focused on IT skills, age, and gender appear contradictory. Research suggests that physicians lack necessary health care industry specific IT skills. The survey instrument used the three aforementioned skills (IGS, IAS, & TS), based on the Learning Skills Profile (LSP), to measure IT skill competency of both entering osteopathic medical students (group 1) and those who graduated medical school (group 2). Careful examination of both groups allowed for such comparison as they had similar gender distribution and Medical College Admission Test (MCAT) scores. A systematic way to measure student learning is to compare student competencies at the beginning and end of their education experience, while time permits, or ensure the two groups are as similar as possible in their demographic characteristics. Data was collected from a sample of 430 students, 230 from Group 1, and 200 from group 2 at a private non-profit university in the southeastern United States. Data was analyzed from 102 participants who took the survey indicating a 24% response rate. Strong reliability was recorded for IGS, IAS, and TS with Cronbach's Alphas of .886, .934, and .937, respectively. Significant difference analysis was done using the non-parametric Mann Whitney U test and skills enhancements were plotted on star-graphs to demonstrate increases, if any, of the measured skills. Overall, IGS and IAS showed significant differences in skill enhancements, while TS did not demonstrate a significant skill enhancement between both groups. Additional attention should be given in current medical schools to enhance the TS of medical students, not just the enhancement of IGS and IAS. Gender testing resulted in a significant difference between the groups, while age did not. Limitations for the study were that both groups were surveyed during the same year from one osteopathic medical school. Future suggestions are presented.

The main goal of this work was to create an XSLT Benchmark, to compare available XSLT processors. At first, we determined main criteria of XSLT processors (price, correctness, speed, memory usage, support, OS and UX), which we discussed. Next, we summarized existing XSLT processors, described individual types of processors (program, library and browser) and measured criteria price, support, OS} and UX. We had to create appropriate test environment to measure criteria correctness, speed and memory usage. Created program XSLT Benchmarking is one of the main benefits of this work. The program also includes tests. We created tests based on 5787 downloaded XSLT files. Tests were created based on the analysis of these files. Firstly, we researched common features of XSLT files (used elements, used XSLT versions, output format etc.). Secondly, we researched focuses of their usages. Next big benefit of this work is a summary of results of tests from different views. We discussed other criteria and features of tested processors. Finally, we summarized possible extensions of our program and also of our analysis.

The problem of liquid loading is common for all gas producing wells and should be identified and solved for efficient gas production. Production engineers and operators need to choose the best solution possible, one that is cost effective and also efficient in doing the job. The plunger lift operation is a cost-effective solution to this liquid loading problem and also is efficient in increasing the gas production. However, the current understanding of plunger lift operation has used field experience and some previous models that have restrictive assumptions which might not be applicable for all plunger lift installations. This research proposes a new plunger lift model that overcomes some of the limiting assumptions of earlier models by using additional data available in the form of pressure and temperature from new technology like smart plunger. The model is based on fundamental principles of mass conservation and pressure balance, and uses the smart plunger data as input. The implementation of the model is carried out in user-friendly and easily accessible software like Excel VBA (Visual Basic Applications). The model predicts the plunger velocity, plunger position and annulus liquid level during an upward travel of the plunger in an onshore gas well in East Texas. The results of model implementation in VBA show the importance of fluid properties for the model, apart from indicating that the model is optimized for the given set of input data. The model developed in this research considers only pressure drop due to gravitational effects, and thus provides a scope for improvement in modeling the plunger lift dynamics by adding frictional and acceleration components. This research also provides recommendations for future work that can be carried out on plunger lift modeling using smart plungers.

<p dir="ltr">We studied the relationship between subsample size and the accuracy of resulted estimation under big data setup.</p><p dir="ltr">We also proposed a novel approach to the construction of confidence intervals based on improved concentration inequalities.</p><p dir="ltr">Lastly, we studied irregular change-point models using free-knot splines.</p>

Thesis (PhD (Physics))--University of Stellenbosch, 2010. / ENGLISH ABSTRACT: The (p, pα) reaction on 12C was investigated experimentally using polarized incident protons of 100 MeV. Coincident data, which were obtained at ten quasifree angle pairs for proton angles ranging from 25◦ to 110◦, were analyzed in terms of the distorted-wave impulse approximation (DWIA). Calculated energy-sharing cross section and analyzing power distributions reproduce the data reasonably well. The observed agreement allows the extraction of distorted momentum distributions from experimental data. These distributions are very consistent over a wide range of angle pairs at which cross section energy-sharing distributions vary considerably. Since measurements of analyzing powers were made, spin-orbit distortions were included in the DWIA calculations. The effects of spin-orbit distortions were found to be very small near zero recoil momentum and did not destroy the validity of the factorization approximation where the two-body p-α cross section enters as a multiplicative factor in the three-body (p, pα) cross section expression. Spectroscopic factors derived from the data are fairly consistent with the trend of the theoretical predictions. Analyzing power data also follow the trend of free p-4He scattering data, and comparisons with DWIA predictions are in reasonable agreement. The theory reproduces also very well analyzing power angular distributions of the projectile-cluster two-body scattering at large angular momentum of the residual nucleus. This indicates that a quasifree knockout mechanism dominates the reaction. The two-body interaction response between the projectile and the α cluster was found to resemble the scattering of protons from a free α particle to a remarkable degree, the present results strongly imply the existence of preformed α clusters in 12C. / AFRIKAANSE OPSOMMING: Die (p, pα) reaksie op 12C is eksperimenteel ondersoek deur middel van gepolariseerde protone met n invalsenergie van 100 MeV. Ko¨ınsidensie data, wat verkry is by tien kwasievrye hoekpare, met proton hoeke tussen 25◦ en 110◦, is geanaliseer in terme van die vervormde-golf-impuls-benadering (DWIA). Die berekende energie-verdeelde kansvlak en analiseervermo¨e verspreidings reproduseer die data redelik goed. Die waargenome ooreenstemming maak dit moontlik om vervormde momentumverdelings uit die eksperimentele data te verkry. Aangesien analiseervermo¨e metings gedoen is, is spin-baan wisselwerking by die DWIA berekenige ingesluit. Die bydra as gevolg van spin-baan wisselwerking blyk baie klein te wees naby nul terugslag momentum en het nie die geldigheid van die faktoriseringsbenadering, waartydens die twee-deeltjie, p-α kansvlak as ’n vermenigvuldigingsfaktor in die uitdrukking vir die drie-deeltjie (p, pα) kansvlak verskyn, vernietig nie. Spektroskopiese faktore wat uit die data herlei is, is redelik konsistent met die verloop van die teoretiese voorspellings. Analiseervermo¨e data volg ook die verloop van die vrye p-4He verstrooiings-data en vergelyk redelik goed met DWIA voorspellings. Die teorie reproduseer ook die hoekverdelings in die analiseervermo¨e van die twee-deeltjie projektiel-bondel verstrooiing by groot hoekmomentum waardes vir die oorblywende kern baie goed. Dit dui daarop dat ’n kwasie-vrye uitslaanmeganisme die reaksie domineer. Die twee-deeltjie wisselwerkingsgedrag tussen die projektiel en die α-bondel toon sterk ooreenkomste met die verstrooiing van protone vanaf ’n vrye α-deeltjie. Die huidige resultate lewer sterk bewyse vir die bestaan van voorafgevormde α-bondels in 12C.

Internet expansion opened up a new business market with a very low barrier to entry. This make online trading as very popular and competitive. The aim of this work is to describe the business on the Internet, especially the stage of promotion, campaign measurement and interpretation. The first chapter deals with the description of tools available for online promotion, technological options for the measurement and subsequent interpretation the results of business activities. The theoretic section describes the entire process on the case, prosperous and rapidly growing international company Pixmac, including detailed descriptions of setup of the analytic system, selection of promotional tools and ROI calculations for optimization and campaign settings.

Today, the average computer user has undoubtedly encountered the PDF format while handling electronic documents. Due to its wide-spread popularity and feature richness, PDF documents are commonly utilized by attackers in order to infect systems with malware. This thesis will present The Extendable Guideline for Analysing Malicious PDF Documents. This work will establish the foundation of the guideline and populate it with a part of the analysis process. The guideline relies on earlier published material in the topic. It is a practical guideline that is followed by the use of a flowchart and can be utilized by an analyst in order to determine if a PDF document is malicious or not. It provides technical background information, suitable analysis techniques, and tools. The guideline structure was developed by using sequential thinking in combination with the divide and conquer paradigm. The thesis will also elucidate commonly applied techniques that are used by malicious PDF authors in order to infect systems, evade detection, and distribute their malicious documents. A commonly utilized function in PDF documents are the JavaScript feature. There are a wide range of other features that are targeted by malicious PDF authors, but they are more rarely encountered. PDF documents are often distributed by attackers by sending them as an attachment in an email, or storing the document on a web server.

<p>An experiment measuring the analyzing power A_y(&theta;) for neutron&ndash;helium-3 (<italic>n</italic>-<super>3</super>He) elastic scattering over broad angular distributions for a range of incident neutron energies from 1.60 to 5.54 MeV has been conducted at the Triangle Universities Nuclear Laboratory. These measurements represent an effort to resolve the long-standing discrepancy between experiment and theory in low-energy three-nucleon analyzing powers, through the evaluation of analyzing powers in the four-nucleon systems, which are expected to exhibit sensitivities not accessible with fewer nucleons. The present work is described in terms of the experimental setup and data reduction techniques; a comparison of the results with rigorous calculations using both nucleon-nucleon and, as recently has become available, three-nucleon potential models is presented. While a discrepancy between calculation and measurement was observed, at low energies substantially better agreement was achieved than in related measurements of the proton&ndash;helium-3 (<italic>p</italic>-<super>3</super>He) analyzing power, suggesting a sizeable dependence on isospin in the four-nucleon systems.</p> / Dissertation