Master of Science / Department of Statistics / Paul Nelson / Effect size is a concept that was developed to bridge the gap between practical and statistical significance. In the context of completely randomized one way designs, the setting considered here, inference for effect size has only been developed under normality. This report is a simulation study investigating the robustness of nominal 0.95 confidence intervals for effect size with respect to departures from normality in terms of their coverage rates and lengths. In addition to the normal distribution, data are generated from four non-normal distributions: logistic, double exponential, extreme value, and uniform. The report discovers that the coverage rates of the logistic, double exponential, and extreme value distributions drop as effect size increases, while, as expected, the coverage rate of the normal distribution remains very steady at 0.95. In an interesting turn of events, the uniform distribution produced higher than 0.95 coverage rates, which increased with effect size. Overall, in the scope of the settings considered, normal theory confidence intervals for effect size are robust for small effect size and not robust for large effect size. Since the magnitude of effect size is typically not known, researchers are advised to investigate the assumption of normality before constructing normal theory confidence intervals for effect size.

O objetivo deste trabalho e determinar a solução de algumas equações de coeficientes intervalares. Este estudo utiliza uma Teoria das Aproximações Intervalares, a qual foi descrita por [ACI91]. Nesta teoria a igualdade para intervalos e substituída pela relação de aproximação . Esta substituição deve-se ao fato da igualdade utilizada na Teoria Clássica dos Intervalos para resolução de equações de coeficientes intervalares não apresentar uma solução satisfatória, visto que a solução encontrada não contem todas as soluções das equações reais que compõe a equação intervalar. Pela substituição da igualdade intervalar por uma relação de aproximação é possível determinar a solução de equações de coeficientes intervalares, de maneira que esta solução contenha todas as possíveis soluções das equações reais pertencentes a equação intervalar. Apresenta-se alguns conceitos básicos, bem como analisa-se algumas propriedades no espaço solução ( /(R), +, •, C, 1). São representadas graficamente diferentes tipos de funções neste espaço intervalar, com os objetivos de obtenção da imagem, caracterização da solução e identificação gráfica da região de solução (ótima e externa), para cada tipo de função. Como a representação de intervalos de /(R) esta determinada num semiplano de eixos X - X+, onde X - representa o extremo inferior de cada intervalo e X+ representa o extremo superior dos intervalos, apresenta-se o espaço intervalar estendido /(R). Neste espaço intervalar estão definidos os intervalos não-regulares, representados no outro semi-piano de eixos X - X+ Em /(R) serão apresentados alguns conceitos fundamentais, assim como operações aritméticas e algumas considerações referentes aos intervalos não-regulares. No espaço intervalar /(R) e possível resolver equações de coeficientes intervalares de maneira análoga a resolução de equações reais no espaço real, pois este espaço intervalar possui a estrutura semelhante a de um corpo. Com isto apresenta-se a solução de equações de coeficientes intervalares lineares, obtida diretamente, assim como determina-se a Formula de Bascara Intervalar para resolução da Equação Quadrática Intervalar. Para funções que possuem grau maior que 2 apresenta-se alguns métodos iterativos intervalares, tais como o Método de Newton Intervalar, o Método da Secante Intervalar e o Método híbrido Intervalar, que permitem a obtenção do intervalo solução para funções intervalares. Por fim apresenta-se alguns conceitos básicos no espaço intervalar matricial M„,„(/(R)), bem como apresenta-se alguns métodos diretos para resolução de sistemas de equações lineares intervalares. / The aim of this work is to determine the solution set of some Equations of Interval Coefficients. The study use a Theory of Interval Approximation. The begining of this theory was described by [ACI91]. In this theory the equality for intervals is replaced by an approximation relation. When we make use of that relation to solve interval equations, it's possible to obtain an optimal solution, i.e., to get an interval solution that contain all of real solutions of the real equations envolved in the interval equation. By using the equality of Classical Interval Theory for solving interval equations we can not get an optimal solution, that is, the interval solution in the most of equations not consider some real solutions of real equations that belong to the interval equation. We present some basic concepts and analyse some properties at the interval space (1(R), E, -a x , 1). Different kind of functions are showed in this space in order to obtain the range, the solution caracterization and the graphic identification of the optimal and external solution region, for each kind of function. The representation of intervals in /(R) is determined in a half plane of axes X - , X+, where X - represent the lower endpoint and X+ represent the upper endpoint of the intervals. The nonregular intervals are defined in /(R), which are determined in an other half plane. In this interval space are presenting some specific concepts, as well as arithmetical operations and some remarks about nonregular intervals. The interval space (1(R), +, •, C, Ex , 1) have a similar structure to a field, so it's possible to solve interval coefficients equations analogously as to solve real equations in the real space. We present the solution of linear interval equations and we determine an interval formula to solve square interval equation. We present some intervals iterated methods for functions that have degree greater than 2 that allow to get an interval solution of interval functions. Finally we show some basic concepts about the interval matrix space Af,„„(IR)) and present direct methods for the resolution of linear interval sistems.

In homicide cases, suitcases provide concealment and ease of transport of a body and reduce the likelihood of detection. When estimating a minimum postmortem interval (mPMI), it is necessary to understand the taphonomic processes that occur when a body is concealed within a suitcase. In this study, the experimental carcasses consisted of pig (Sus scrofa Linnaeus) heads concealed within either hard shell or fabric suitcases whereas control pig heads were decomposed naturally. All control pigs heads attracted blowflies (Calliphoridae) immediately and mummified within days, whereas the experimental pigs heads had an oviposition delay of approximately 72 hours and decomposed resulting in skeletonization. Differences in the arthropod species present varied between the experimental and control pigs heads. Temperature comparisons revealed significant (p <0.001) differences between the experimental and control pigs heads. How a corpse decomposes when a suitcase is used for body disposal must be considered when estimating the mPMI.

Introducción: La desnutrición crónica repercute en el crecimiento y desarrollo del niño menor de 5 años. Investigaciones afirman que la lactancia materna exclusiva junto con otras estrategias, como un intervalo de nacimiento mayor a 24 meses, favorecen al adecuado crecimiento del niño. Nuestro objetivo fue determinar la asociación entre el intervalo de nacimiento y el z-score de talla / edad en niños menores de 5 años, en presencia o ausencia de lactancia materna exclusiva. Materiales y métodos: Se utilizó la base de datos de la ENDES 2017. Para la asociación se trabajó con el intervalo de nacimiento y el z-score de la talla / edad, de niños menores de 5 años y se empleó la variable de estratificación lactancia materna exclusiva. Se aplicó un modelo de regresión lineal para estimar la asociación del z-score de la talla / edad e intervalo de nacimiento según presencia o ausencia de lactancia materna exclusiva. Resultados: Los resultados de la presente investigación sugieren que existe asociación entre el intervalo de nacimiento y el z-score de la talla / edad. Los niños con intervalo de nacimiento menor o igual a 24 meses tienen menor z-score de talla / edad en comparación a los niños con un intervalo de nacimiento mayor a 24 meses. Conclusión: Los niños con intervalo de nacimiento menor o igual a 24 meses se asociaron con un menor z-score de talla / edad independientemente si han lactado o no de forma exclusiva. / Background: Chronic malnutrition has an impact on the growth and development of children under 5 years of age. Research states that exclusive breastfeeding along with other strategies, such as a birth interval greater than 24 months, favor the development and proper growth of the child. Our objective was to determine the association between the birth interval and the z-score of height for age in children under 5 years of age, in the presence or absence of exclusive breastfeeding Methods: The database of the ENDES 2017 was used. For the association we worked with the birth interval and the z-score of the height for the age of each child under 5 years old and the stratification variable exclusive breastfeeding was used. A linear regression model was applied to estimate the association of height z-score for age and birth interval according to the presence or absence of exclusive breastfeeding. Results: The results of the present investigation suggest that there is an association between the birth interval and the z-score of the height for the age. Children with a birth interval of less than 24 months have a lower z-score of height for age compared to children with a birth interval greater than 24 months. / Tesis

A la fin des années 1880, le langage musical de Claude Debussy, influencé à la fois par Wagner, le symbolisme, et la poétique de Baudelaire, se transforme en profondeur.Le système tonal, alors unique horizon d'attente (Jauss) des institutions musicales européennes, est repoussé à l'arrière plan des mélodies du compositeur, par des associations avec des thèmes poétiques comme les souvenirs, l'humour, et l'enfermement.Cette thèse se concentre sur une analyse de ces transformations, et sur une comparaison de ce langage musical neuf avec celui des Serres chaudes d'Ernest Chausson, cycle de mélodies qui partage pourtant les mêmes influences poétiques et musicales. / In the late 1880's, Claude Debussy's musical language, influenced by Wagner, symbolism and Baudelaire's poetry, deeply transforms itself.The tonal system, still regarded as the unique horizon of expectations (Jauss) of European musical institutions, is pushed to the background in the composer's melodies by his characteristic associations with poetical themes like memories, humor, and reclusion.This thesis focuses on an analysis of these transformations and on a comparison of this new musical language with Ernest Chausson's Serres chaudes, which still shares the same poetic and musical influences.

The term quote semantic query optimization quote (SQO) denotes a methodology whereby queries against databases are optimized using semantic information about the database objects being queried. The result of semantically optimizing a query is another query which is syntactically different to the original, but semantically equivalent and which may be answered more efficiently than the original. SQO is distinctly different from the work performed by the conventional SQL optimizer. The SQL optimizer generates a set of logically equivalent alternative execution paths based ultimately on the rules of relational algebra. However, only a small proportion of the readily available semantic information is utilised by current SQL optimizers. Researchers in SQO agree that SQO can be very effective. However, after some twenty years of research into SQO, there is still no commercial implementation. In this thesis we argue that we need to quantify the conditions for which SQO is worthwhile. We investigate what these conditions are and apply this knowledge to relational database management systems (RDBMS) with static schemas and infrequently updated data. Any semantic query optimizer requires the ability to reason using the semantic information available, in order to draw conclusions which ultimately facilitate the recasting of the original query into a form which can be answered more efficiently. This reasoning engine is currently not part of any commercial RDBMS implementation. We show how a practical semantic query optimizer may be built utilising readily available semantic information, much of it already captured by meta-data typically stored in commercial RDBMS. We develop cost models which predict an upper bound to the amount of optimization one can expect when queries are pre-processed by a semantic optimizer. We present a series of empirical results to confirm the effectiveness or otherwise of various types of SQO and demonstrate the circumstances under which SQO can be effective.

The development and improvement of methods used for the estimation of the postmortem interval (PMI) is a common area of research in forensic science. This research was conducted to physically and chemically analyze pig carcass decomposition on a soil surface using conventional and newly developed methods for the potential use in estimating the PMI. Photographs of pig carcasses decomposing on forested and open land were scored using a decomposition scoring system and decomposition scores were related to accumulated degree days (ADD). Overall, the ADD values were significantly different for the two groups of carcasses; however, the ADD values for the onset of each score demonstrated more similarity between groups. Decomposition scoring results also indicated that refinements must be made to the calculation of ADD to allow for a meaningful comparison of pig and human decomposition. The decomposition of pig carcasses altered the water content, pH and fatty acid content of soil. The fatty acids, myristic, palmitic, palmitoleic, stearic and oleic acids were successfully extracted and analyzed from decomposition soil. Palmitic, stearic and oleic acids were the most abundant fatty acids detected whilst the levels of myristic and palmitoleic acids were negligible in comparison. A three peak fatty acid cycle was also observed for each fatty acid. Variations in soil pH and fatty acid content of decomposition soil have the potential to indicate the presence of a decomposition site. Furthermore, a nonlinear diffusion model was developed to predict the development of the cadaver decomposition island (CDI) in soil over time. The simulation of the model indicated that the diffusion model has the potential to generate PMI estimations for early stages of decomposition by corresponding the effective radius of the CDI to a particular time point. The general findings of this research indicate that more accurate methods for PMI estimations can potentially be developed with further research. / UOIT

In statistics it is of interest to find a better interval estimator of the absolute mean deviation. In this thesis, we focus on using the jackknife, the adjusted and the extended jackknife empirical likelihood methods to construct confidence intervals for the mean absolute deviation of a random variable. The empirical log-likelihood ratio statistics is derived whose asymptotic distribution is a standard chi-square distribution. The results of simulation study show the comparison of the average length and coverage probability by using jackknife empirical likelihood methods and normal approximation method. The proposed adjusted and extended jackknife empirical likelihood methods perform better than other methods for symmetric and skewed distributions. We use real data sets to illustrate the proposed jackknife empirical likelihood methods.

Le calcul ensembliste par analyse par intervalles et ses applications

This paper studies the interval estimation of three discrete distributions: thebinomial distribution, the Poisson distribution and the negative-binomialdistribution. The problem is the chaotic behavior of the coverage probabilityfor the Wald interval. To solve this problem, alternative confidence intervals areintroduced. Coverage probability and expected length are chosen to be thecriteria evaluating the intervals.In this paper, I firstly tested the chaotic behavior of the coverageprobability for the Wald interval, and introduced the alternative confidenceintervals. Then I calculated the coverage probability and expected length forthose intervals, made comparisons and recommended confidence intervals forthe three cases. This paper also discussed the relationship among the threediscrete distributions, and in the end illustrated the applications on binomialand Poisson data with brief examples.