One of the most useful and best known goodness of fit test is the Kolmogorov one-sample test. The assumptions for the Kolmogorov (one-sample test) test are: 1. A random sample; 2. A continuous random variable; 3. F(x) is a completely specified hypothesized cumulative distribution function. The Kolmogorov one-sample test has a wide range of applications. Knowing the effect fromusing the test when an assumption is not met is of practical importance. The purpose of this research is to analyze the robustness of the Kolmogorov one-sample test to sampling from a finite discrete distribution. The standard tables for the Kolmogorov test are derived based on sampling from a theoretical continuous distribution. As such, the theoretical distribution is infinite. The standard tables do not include a method or adjustment factor to estimate the effect on table values for statistical experiments where the sample stems from a finite discrete distribution without replacement. This research provides an extension of the Kolmogorov test when the hypothesized distribution function is finite and discrete, and the sampling distribution is based on sampling without replacement. An investigative study has been conducted to explore possible tendencies and relationships in the distribution of Dn when sampling with and without replacement for various parameter settings. In all, 96 sampling distributions were derived. Results show the standard Kolmogorov table values are conservative, particularly when the sample sizes are small or the sample represents 10% or more of the population.

The investigation sought to examine resistance to extinction (Rn) as a function of previous experience with downward shifts in reward magnitude. It was suggested that previous research conducted within the framework of the Spence-Amsel frustration hypothesis and the sequential hypothesis failed to administer sufficient postshift trials to adequately establish the relationships that may exist. Under one condition, four groups of rats received twenty extinction trials following forty postshift trials. Under another condition, four groups were extinguished following eighty postshift trials. An inverse magnitude of reward effect occurred in the preshift phases, however, which prevented an adequate analysis of either the shift or the Rn data, This unexpected effect was discussed within the framework of Black's incentive-motivation interpretation of reinforcement.

This thesis makes fundamental computational and statistical advances in testing and estimation, making critical progress in theory and application of classical statistical methods like classification, regression and hypothesis testing, and understanding the relationships between them. Our work connects multiple fields in often counter-intuitive and surprising ways, leading to new theory, new algorithms, and new insights, and ultimately to a cross-fertilization of varied fields like optimization, statistics and machine learning. The first of three thrusts has to do with active learning, a form of sequential learning from feedback-driven queries that often has a provable statistical advantage over passive learning. We unify concepts from two seemingly different areas—active learning and stochastic firstorder optimization. We use this unified view to develop new lower bounds for stochastic optimization using tools from active learning and new algorithms for active learning using ideas from optimization. We also study the effect of feature noise, or errors-in-variables, on the ability to actively learn. The second thrust deals with the development and analysis of new convex optimization algorithms for classification and regression problems. We provide geometrical and convex analytical insights into the role of the margin in margin-based classification, and develop new greedy primal-dual algorithms for non-linear classification. We also develop a unified proof for convergence rates of randomized algorithms for the ordinary least squares and ridge regression problems in a variety of settings, with the purpose of investigating which algorithm should be utilized in different settings. Lastly, we develop fast state-of-the-art numerically stable algorithms for an important univariate regression problem called trend filtering with a wide variety of practical extensions. The last thrust involves a series of practical and theoretical advances in nonparametric hypothesis testing. We show that a smoothedWasserstein distance allows us to connect many vast families of univariate and multivariate two sample tests. We clearly demonstrate the decreasing power of the families of kernel-based and distance-based two-sample tests and independence tests with increasing dimensionality, challenging existing folklore that they work well in high dimensions. Surprisingly, we show that these tests are automatically adaptive to simple alternatives and achieve the same power as other direct tests for detecting mean differences. We discover a computation-statistics tradeoff, where computationally more expensive two-sample tests have a provable statistical advantage over cheaper tests. We also demonstrate the practical advantage of using Stein shrinkage for kernel independence testing at small sample sizes. Lastly, we develop a novel algorithmic scheme for performing sequential multivariate nonparametric hypothesis testing using the martingale law of the iterated logarithm to near-optimally control both type-1 and type-2 errors. One perspective connecting everything in this thesis involves the closely related and fundamental problems of linear regression and classification. Every contribution in this thesis, from active learning to optimization algorithms, to the role of the margin, to nonparametric testing fits in this picture. An underlying theme that repeats itself in this thesis, is the computational and/or statistical advantages of sequential schemes with feedback. This arises in our work through comparing active with passive learning, through iterative algorithms for solving linear systems instead of direct matrix inversions, and through comparing the power of sequential and batch hypothesis tests.

The empirical data indicates that Generation Y generally has a very high average financial literacy. The rationality they exhibit in the mastery of financial information and tools does not affect their savings and investments. Instead, the investment behavior is given in expression by their character, but also demographic basis.

It is widely acknowledged that under situations of heightened pressure, many expert athletes suffer from performance decrements. This phenomenon has been termed ‘choking under pressure’ and has been the subject of extensive research in sport psychology. Despite this attention, gaps in the literature remain leaving opportunities for further advancements in knowledge about the phenomenon, particularly in relation to its underlying processes and the development of appropriate interventions that can be adopted in order to alleviate, or even prevent choking. The present programme of research, in general terms, aimed to develop and test the efficacy of an intervention tool, based on priming, to alleviate choking under pressure. It was acknowledged that such a tool should be matched to the mechanisms that underlie the choking process and although an abundance of research has provided valuable information about these mechanisms, it was identified that there still remains a lack of consensus regarding the most appropriate explanatory theory. Therefore the initial study in this thesis aimed to provide further insight into the processes that govern choking by examining accounts from elite international swimmers of their experiences of performing under high levels of pressure. The results provided further support for the postulation that choking under pressure occurs as a result of a combination of conscious processing hypothesis (Masters, 1992) and processing efficiency theory (Eysenck & Calvo, 1992) and that an optimum level of skill-focused attention is beneficial to performance. The following studies utilised this information as well as that of the existent theories of choking, to develop and examine an effective priming based intervention tool (a scrambled sentence task). Specifically, Studies 2, 3 and 4 examined the amount of residual working memory available after activation of the prime, the optimisation of the priming task and the efficacy of the tool in promoting performance under high pressure respectively. Results revealed support for the efficacy of the tool in reducing online skill-focused attention and promoting performance under both low- and high-pressure conditions. Finally, the general themes that emerged throughout the whole programme of study are discussed, as well as the limitations and recommendations for future research. Implications for coaches, athletes and practitioners are also presented.

Adolescent obesity has increased exponentially over the past three decades in the United States. In response, behavioral interventions have been developed and implemented to address this epidemic; however, treatment adherence is often suboptimal. Motivational interviewing (MI) is a directive, person centered approach to reducing patient ambivalence about change, which has been shown to increase engagement in obesity interventions. The current study investigated the underlying process of MI by exploring two different, but related pathways that explain how change happens (e.g., the technical and relational hypotheses) in the context of a multidisciplinary obesity intervention with African American adolescent girls (N = 30). Results demonstrated that MI-consistent skills were associated with client “change talk,” or language consistent with their desire, ability, reasons, need, taking steps or commitment to change. Specifically, clinician reflection of client change language and support of client autonomy were especially important for increasing change language. Furthermore, client language related to change was associated with three-month treatment adherence, and increased fruit and vegetable intake. However, the clinician’s ability to embody the MI “spirit” was not related to either client language or three-month treatment adherence. Findings suggest that the technical aspect of MI helps explain this approach's effects on treatment adherence among African American girls with obesity.

Title: Permutation Tests of Statistical Hypotheses Author: Zdeněk Veselý Department: Department of Probability and Mathematical Statistics Supervisor: prof. RNDr. Jana Jurečková DrSc., Department of Probability and Mathematical Statistics Abstract: This thesis presents permutation tests concept. Permutation test is demonstrated as response to testing problems where it is inconvenient to make any deeper presumptions on data probability distribution. For some of these problems it is even the only exact solution. The construction of permutation test is described in the thesis as well as approach to search of the most powerful tests to specific alternatives. In the second part of the thesis there are comparisons of powers of parametric, permutation and rank test using simulations. The result is that power of parametric and permutation test are very similar most of the times and that confirms that permutation tests are useful tool for praxis. Keywords: Permutation tests, Exact tests, Hypothesis testing, Power of tests

Dans le cadre de nombreuses règlementations, l’établissement de stratégies de suivi des populations de cétacés est devenu un besoin prioritaire. Les principales caractéristiques d’une stratégie de suivi sont la signification écologique, la crédibilité statistique et le rapport qualité/coût élevé. La mise en place d’indicateurs permet de réduire les coûts de suivi, et propose un outil de communication entre scientifiques et gestionnaires. Un indicateur est une donnée vérifiable et mesurable qui renseigne sur plus qu’elle-même. La collecte des données d’échouages est peu coûteuse, mais l’absence de stratégie d’échantillonnage des populations en mer est une entrave à leur utilisation dans le cadre de stratégies de suivi. L’objectif de cette thèse est de développer des indicateurs des populations de cétacés à partir des données d’échouages. Quatre indicateurs ont été développés au cours de cette thèse. La part des cétacés échoués par rapport aux cétacés morts en mer a été estimée. Les séries temporelles d’échouages ont été affranchies des conditions de dérive, par l’établissement de l’hypothèse nulle (hypothèse d’uniformité spatiale et temporelle des cétacés morts en mer) et constitue l’anomalie d’échouage. La cartographie de la mortalité de cétacés a été réalisée, et détermine le nombre de dauphins morts en mer, indépendamment de leur probabilité de s’échouer. Enfin la distribution de mortalité observée par rapport à la distribution attendue des cétacés morts sous l’hypothèse nulle a été calculée. Elle permet d’identifier des zones de fortes mortalité ou abondance relative. Ces quatre indicateurs pourront être intégrés à différents plans de gestion, nationaux ou internationaux. / Under national and international regulations, the establishment of cetacean population monitoring strategies became a priority. Monitoring main expected characteristics are the ecological significance, the statistical credibility and the cost-effectiveness. The use of indicators reduces monitoring cost, and constitutes communication tool between scientists and policy-makers. Indicators are measurable and verifiable data, which inform on more than themselves. The collect of stranding data is cost-efficient, but the lack of sampling strategy remains a hindrance to their use in context of monitoring strategy. The aim of this study is to develop cetacean population indicators based on stranding data.Four indicators were developed during this work. The proportion of stranded cetaceans compared to dead cetaceans at sea was estimated. Stranding long term time series were freed from drift conditions by the construction of the null hypothesis (that is a hypothesis of spatial and temporal uniformity of dead cetaceans). It constituted the stranding anomaly. The cetacean mortality cartography estimated the number of dead cetaceans at sea, independently of the stranding probability. Finally, cetacean mortality anomaly was calculated as the difference between mortality areas of stranded cetaceans and theoretical distribution of dead cetaceans predicted to strand under the null hypothesis. This anomaly identified high mortality or high relative abundance areas. These indicators could be integrated to many national and international management strategies.

Aggression is a phenomenon which acquired a substantial share on attention of biological literature, especially since the beginning of 20th century. In this thesis, various approaches to the problem from the field of biology are summarized. Intensified attention is dedicated to the theoretical concepts of aggression, especially the concepts consulting the problem of aggression from the evolutionary point of view and to the research on the field of hormonal regulation of aggression, stressing out the role of the hormone testosterone, and to particular hypothesis, which take account on its mutual relationship with behavior. The empirical part of the thesis is dedicated to verification of premises of the challenge hypothesis according to the relationship between testosterone and aggressive behavior. Significant findings of the empirical part refer to the existence of a link between basal level of testosterone and aggressive tendencies and to the existence of a negative link between basal level of testosterone and its increase in reaction to a competitive situation.

Hurricanes can temporarily disrupt seasonal patterns of fish assemblage change or result in permanent changes in fish assemblages. I studied the effects of two hurricanes on fish assemblages at the Chandeleur Islands and the possible influence that storm-generated tidal channels may have on the composition of local fish assemblages. I also compared recently collected data to historic ichthyofaunal survey data collected over thirty years ago at the Chandeleur Islands. Near shore fish assemblages changed the most after hurricanes but changes in species composition were primarily due to increases in abundance and diversity. During July 2007 there was no significant difference between fish assemblages in channel and seagrass habitats, although significant differences among wash-over channels existed. Loss of habitat and the increased intensity and frequency of recent storms may explain why current fish assemblages at the Chandeleur Islands are less diverse (as measured by taxonomic distinctness) than assemblages collected during 1969-1971.