Bootstrapping plays a vital role in the life of small and medium-sized enter-prises. By providing a large variety of financing alternatives bootstrapping ensures the existence of entrepreneurship, even though, too less attention is paid to bootstrapping in the specific literature. Therefore, the master thesis strives to eliminate the gaps in the theory by bringing new insights in the field of bootstrapping.The purpose of the master thesis is to investigate the usage of boot-strapping methods and understand the main motives for using these methods from a sectorial and cross-national point of view.In order to accomplish the purpose, the research has been carried out among travel agencies and tour operators (both belonging to the tourism sector) from Sweden, Latvia and Romania by conducting 11 semi-structured interviews, as well as by collecting 50 internet-mediated questionnaires.Empirical findings were analyzed and interpreted through the 4-category Model that was created for gaining a deeper understanding of the phenom-ena of bootstrapping. Results showed that there is a certain trend in the usage of bootstrapping methods within the tourism sector. Thus, seek-ing out best conditions possible with supplier/s and obtaining payments in advance from customer are recognized to be the most important bootstrap-ping techniques for the travel agencies and tour operators. Moreover, it was found that the main motive for bootstrapping is cost reduction. In addi-tion, the results revealed that there are no significant differences in the usage of bootstrapping methods and in the motivation that lay behind bootstrapping from a cross-national point of view.

The motivation for this thesis came from the provision of a large data set from Saudi Arabia giving anthropometric measurements of children and adolescents from birth to eighteen years of age, with a requirement to construct growth charts. The construction of these growth charts revealed a number of issues particularly in the respect to statistical inference relating to quantile regression. To investigate a range of different statistical inference procedures in parametric quantile regression in particular the estimation of the confidence limits of the ?th (?? [0, 1]) quantile, a number of sets of simulated data in which various error structures are imposed including homoscedastic and heteroscedastic structures were developed. Methods from the statistical literature were then compared with a method proposed within this thesis based on the idea of Silverman's (1986) kernel smoothing. This proposed bootstrapping method requires the estimation of the conditional variance function of the fitted quantile. The performance of a variety of variance estimation methods combined within the proposed bootstrapping procedure are assessed under various data structures in order to examine the performance of the proposed bootstrapping approach. The validity of the proposed bootstrapping method is then illustrated using the Saudi Arabian anthropometric data.

Svensk filmindustri idag finansieras huvudsakligen genom statliga medel. Dock finns det inte tillräckligt med finansiella medel jämfört det antal svenska filmer som produceras varje år. När finansieringen är otillräcklig kan det vara av största vikt att förstå och kunna arbeta med finansiell bootstrapping som verktyg för att kunna slutföra produktionen av filmen. Finansiell bootstrapping är ett begrepp som kan fungera som ett resursanskaffande verktyg för att tillföra resurser till en produktion till en lägre kostnad än marknadspriset, eller till ingen kostnad alls. Denna uppsats syftar till att undersöka om detta verktyg finns inom svensk filmproduktion – och i så fall vilka som använder verktyget och varför. Forskningen avser också att utreda tillvägagångssätt och tankar kring känslor om användandet, både inom kort- och långfilmsproduktioner. Vi har funnit att användning av finansiell boostrapping finns och är generellt sett mer frekvent använt i produktioner som inte är fullt finansierade. Förekomsten av användandet är oftast vanligare i kortfilm än i långfilm, men att det existerar inom båda formerna av filmproduktion samt att verktyget används av olika befattningshavare. Utsagorna om känslor kring användandet av finansiell bootstrapping är tvetydiga – det uppstår enligt respondenterna ofta positiva känslor när det finns en ömsesidig vinning mellan de berörda avtalsparterna, men också negativa känslor när utbytet av överenskommelsen är mest gynnsamt för den ena parten. / Contemporary Swedish film industry is financed mainly by government funding. However, there is potentially not enough funding in comparison to how many films are produced in Sweden every year. When funding is inadequate it could be paramount to master the concept and instrument of financial bootstrapping in order to complete a film. Financial bootstrapping is a term and tool regarding the acquirement of resources, albeit to a lower cost than market value or to no costs at all. This essay aims to investigate whether this phenomenon exist within the Swedish film industry, if so - who is using this tool and why. The research also intends to chart the approach and feelings of users both in short film as well as full feature film. We have found that utilization exists and is generally more frequent amongst films that are not fully financed. The occurrence is often more common in short films than in feature films; also it exists within the different levels of film production and is used by various positions. Thoughts regarding the practice of financial bootstrapping are ambiguous; there are positive approaches when there is a mutual gain between the relevant participants, but also, negative approaches when the benefits from the collaboration are divided unfairly.

We investigate the use of machine learning to create effective heuristics for single-agent search. Our method aims to generate a sequence of heuristics from a given weak heuristic h{0} and a set of unlabeled training instances using a bootstrapping procedure. The training instances that can be solved using h{0} provide training examples for a learning algorithm that produces a heuristic h{1} that is expected to be stronger than h{0}. If h{0} is so weak that it cannot solve any of the given instances we use random walks backward from the goal state to create a sequence of successively more difficult training instances starting with ones that are guaranteed to be solvable by h{0}. The bootstrap process is then repeated using h{i} instead of h{i-1} until a sufficiently strong heuristic is produced. We test this method on the 15- and 24-sliding tile puzzles, the 17- , 24- , and 35-pancake puzzles, Rubik's Cube, and the 15- and 20-blocks world. In every case our method produces heuristics that allow IDA* to solve randomly generated problem instances quickly with solutions very close to optimal. The total time for the bootstrap process to create strong heuristics for large problems is several days. To make the process efficient when only a single test instance needs to be solved, we look for a balance in the time spent on learning better heuristics and the time needed to solve the test instance using the current set of learned heuristics. %We use two threads in parallel, We alternate between the execution of two threads, namely the learning thread (to learn better heuristics) and the solving thread (to solve the test instance). The solving thread is split up into sub-threads. The first solving sub-thread aims at solving the instance using the initial heuristic. When a new heuristic is learned in the learning thread, an additional solving sub-thread is started which uses the new heuristic to try to solve the instance. The total time by which we evaluate this process is the sum of the times used by both threads up to the point when the instance is solved in one sub-thread. The experimental results of this method on large search spaces demonstrate that the single instance of large problems are solved substantially faster than the total time needed for the bootstrap process while the solutions obtained are still very close to optimal.

A growing number of applications require dictionaries of words belonging to semantic classes present in specialized domains. Manually constructed knowledge bases often do not provide sufficient coverage of specialized vocabulary and require substantial effort to build and keep up-to-date. In this thesis, we propose a semi-supervised approach to the construction of domain-specific semantic lexicons based on the distributional similarity hypothesis. Our method starts with a small set of seed words representing the target class and an unannotated text corpus. It locates instances of seed words in the text and generates lexical patterns from their contexts; these patterns in turn extract more words/phrases that belong to the semantic category in an iterative manner. This bootstrapping process can be continued until the output lexicon reaches the desired size. We explore employing techniques such as learning lexicons for multiple semantic classes at the same time and using feedback from competing lexicons to increase the learning precision. Evaluated for extraction of dish names and subjective adjectives from a corpus of restaurant reviews, our approach demonstrates great flexibility in learning various word classes, and also performance improvements over state of the art bootstrapping and distributional similarity techniques for the extraction of semantically similar words. Its shallow lexical patterns also prove to perform superior to syntactic patterns in capturing the semantic class of words.

Bootstrapping kan användas som komplement eller substitut till extern finansiering som metoder för resursanskaffning till lägsta möjliga kostnad. Samtliga respondenter föredrog internt genererade medel framför banklån. Ökning av eget kapital med hjälp av externa finansiärer visade sig vara det minst önskvärda alternativet för kapitalanskaffning. Studien visade att ägartillförda medel och nyttjande av släkt och vänner var de vanligast förekommande metoderna för bootstrapping i de tillfrågade företagen. Av de tillfrågade företagen var de som ansåg sig existera på en riskfylld marknad mer benägna att använda sig av bootstrappingmetoder. / Bootstrapping can be used as an addition or a substitute for external financing as methods for acquisition of resources at lowest possible cost. All of the respondents preferred internally generated finances to bank loans. Financing by issuing equity was the least desirable method. The study showed that owner’s capital and usage of friends and family as cheap labor are the most commonly occurring methods of bootstrapping in the surveyed corporations. Among the surveyed corporations, those operating on unpredictable markets are more likely to use bootstrapping methods.

The purpose of this study was to examine factors supporting college mathematics success. First, effect of a brief high school orientation to mathematical technologies used for college placement testing was examined. Secondly, the voice of participants in this orientation was heard. Finally, bootstrapped orientation data were presented to teachers and instructors of introductory statistics courses as a scaffold to their technological pedagogical content knowledge (TPCK) as these teachers and instructors strive to actively engage students to achieve college mathematics success. Many entering college students are placed into developmental mathematics classes based on scores from college placement assessments that allow extremely limited use of calculating technology and have various time constraints. Students in a rural central Texas 3A high school that were enrolled in Algebra II course were given pre- and post- tests in Arithmetic and Algebra. Each 20-minute test contained 15 mathematical content questions and one qualitative question. The post-test was given approximately a week after the pre-test. During the week, students were provided time to explore review material using only pencil and paper for the arithmetic review, and a four-function calculator on the algebra review questions. Effects of the orientation were analyzed using mean scores, confidence intervals, effect size, and GLM for whole-group and sub-groups. A paired samples t-test was calculated. These effects were discussed. A case study involving participants of the orientation was conducted. Twelve participants were interviewed after each had entered college. Five themes emerged from these interviews: (1) Knowledge of College Mathematics, (2) Technology and Mathematics, (3) Mathematics Tests/Assessments, (4) Teaching and Learning Mathematics, and (5) Mathematical Experiences, Hopes and Dreams. Each theme is discussed. Using Microsoft Excel, bootstrapping is presented to instructors of first year introductory statistics courses in support of student success as instructors’ technological pedagogical content knowledge is developed. A course project demonstrating and developing application of computational technology by bootstrapping confidence intervals at the 95 % level using Microsoft Excel is presented. Data from the orientation were further analyzed in the bootstrapping project. Confidence intervals were empirically calculated from bootstrapped resamples of the mean. The number of resamples used was 250 at each of three levels: Over-sampling, at-sampling, and under-sampling. Graphs of bootstrapped confidence intervals, using the Rule of Eye 4, showed statistically significant differences between pre-test and post-test scores for all pairs of data sets.

No description available.

A growing number of applications require dictionaries of words belonging to semantic classes present in specialized domains. Manually constructed knowledge bases often do not provide sufficient coverage of specialized vocabulary and require substantial effort to build and keep up-to-date. In this thesis, we propose a semi-supervised approach to the construction of domain-specific semantic lexicons based on the distributional similarity hypothesis. Our method starts with a small set of seed words representing the target class and an unannotated text corpus. It locates instances of seed words in the text and generates lexical patterns from their contexts; these patterns in turn extract more words/phrases that belong to the semantic category in an iterative manner. This bootstrapping process can be continued until the output lexicon reaches the desired size. We explore employing techniques such as learning lexicons for multiple semantic classes at the same time and using feedback from competing lexicons to increase the learning precision. Evaluated for extraction of dish names and subjective adjectives from a corpus of restaurant reviews, our approach demonstrates great flexibility in learning various word classes, and also performance improvements over state of the art bootstrapping and distributional similarity techniques for the extraction of semantically similar words. Its shallow lexical patterns also prove to perform superior to syntactic patterns in capturing the semantic class of words.

Changepoint analysis is a useful tool in environmental statistics in that it provides a methodology for threshold detection and modeling processes subject to periodic changes in the underlying model due to anthropogenic effects or natural phenomena. Several applications of changepoint analysis are investigated here. The use of inappropriate changepoint detection methods is first discussed and the need for a simple, flexible, correct method is established and such a method is proposed for the mean-shift model. Data from the Everglades, Florida, USA is used to showcase the methodology in a real-world setting. An extension to the case of time-series data represented via transition matrices is presented as a result of joint work with Matt Williams (Department of Statistics, Virginia Tech) and rainfall data from Kenya, Africa is presented as a case-study. Finally the multivariate changepoint problem is addressed by a two-stage approach beginning with dimension reduction via principal component analysis (PCA). After the dimension reduction step the location of the changepoint in principal component space is estimated and assuming at most one change in a mean-shift setting, all possible sub-models are investigated. / Ph. D.