Treatment–covariate interaction is often used in clinical trials to assess the homogeneity of treatment effects over these subgroups defined by a baseline covariate, which is frequently conducted after primary analysis including all patients is completed. When the endpoint is the time to an event, as in the cancer clinical trials, the Cox proportional hazard model with an interaction term has been used exclusively to test the significance of treatment-covariate interaction in oncology literature. But the proportional hazards assumption may not be satisfied by the data from clinical trials. Although there are several procedures proposed in statistical literature to assess the interaction based on a nonparametric measure of interaction or nonparametric models, some of these procedures do not take into the account of the nature of the data well, while some are very complicated which may have limited their applications in practice. In this thesis, a non-parametric procedure based on the smoothed estimate of Patel–Hoel measure is first derived to test the interaction between the treatment and a binary covariate with censored data. The theoretical distribution of the test statistic of the proposed procedure is derived. The proposed procedure is also evaluated through Monte-Carlo simulations and applications to data from a cancer clinical trial. Jackknifed versions of two test statistics based on nonparametric models are then derived by simplifying these test statistics and applying the jackknife method to estimate their variances. These jackknifed tests are also compared with the smoothed test and other related tests. / Thesis (Ph.D, Mathematics & Statistics) -- Queen's University, 2011-09-27 11:09:28.449

No description available.

No description available.

Tese (doutorado)—Universidade de Brasília, Instituto de Geociências, 2014. / Submitted by Albânia Cézar de Melo (albania@bce.unb.br) on 2014-09-11T13:57:16Z No. of bitstreams: 1 2014_CesarGarciaPavao.pdf: 71280780 bytes, checksum: 27319c0108df7bbcba44ae950d914b01 (MD5) / Approved for entry into archive by Guimaraes Jacqueline(jacqueline.guimaraes@bce.unb.br) on 2014-09-29T12:16:12Z (GMT) No. of bitstreams: 1 2014_CesarGarciaPavao.pdf: 71280780 bytes, checksum: 27319c0108df7bbcba44ae950d914b01 (MD5) / Made available in DSpace on 2014-09-29T12:16:12Z (GMT). No. of bitstreams: 1 2014_CesarGarciaPavao.pdf: 71280780 bytes, checksum: 27319c0108df7bbcba44ae950d914b01 (MD5) / A utilização de metodologias de interpolação espacial de dados está se tornando cada vez mais frequente nas análises geofísicas, em função de que, atualmente, diversos softwares já contêm diversos desses métodos, permitindo análises mais detalhadas. No presente trabalho avaliam-se oito métodos de interpolação espacial, para dados de espessura crustal, provenientes de função do receptor, sísmica e gravimetria, das províncias tectônicas do Brasil, com intuito de confeccionar uma mapa da Moho para essas regiões. Os métodos utlizados foram IDW, Natural Neighbor, Spline, Kriging, Radial Basis Function, Local Polynomial Interpolation, Global Polynomial Interpolation e Topo to Raster. Foram compilados 776 dados de espessura crustal sendo 281 dentro do limite do território brasileiro, os demais foram utilizados para minimizar os erros de borda gerados durante o processo de interpolação. Os dados foram processados através das ferramentas de interpolação localizadas no Spatial Analyst Tools e Geostatistical Wizard do programa ARCGIS 10. Para validação dos modelos foi utilizada a técnica de reamostragem denominada Jackknife, a qual foi de suma importância para normalização dos valores mínimos e máximos calculados e corroborando com a obtenção dos resíduos gerados no processo. Para comparação entre os modelos foram utilizados seis critérios estatísticos, nos quais os modelos Kriging, Topo to Raster e Global Polynomial Interpolation apresentaram os melhores resultados. Os modelos obtidos pelos métodos Spline, Radial Basis Function não honraram os dados de entrada. Já os modelos IDW, Natural Neighbor e Global Polynomial Interpolation geraram anomalias não condizentes com as características físicas da crosta, enquanto o Local Polynomial Interpolation gerou um modelo suave suprimindo informações importantes sobre a crosta. Dentre os mod to Raster foram elos analisados o kriging e o Topoos que melhor fizeram. Com a publicação de novos dados de espessura crustal para o Brasil, modelos poderão ser gerados de forma rápida, prática e eficiente. _________________________________________________________________________ ABSTRACT / The use of methods for data spatial interpolation is becoming increasingly common in geophysical analyses, because of that, currently, many softwares already contain many of these methods, allowing a more detailed analysis. In this paper, we evaluate eight methods of spatial interpolation for data of crustal thickness from Receiver Function, seismic and gravity data of Brazil’s tectonic provinces, aiming to prepare a map of the Moho for these regions. The methods used were IDW, Natural Neighbor, Spline, Kriging, Radial Basis Function, Local Polynomial Interpolation, Global Polynomial Interpolation and Topo to Raster. There were 776 crustal thickness data compiled with 281 being within the limits of the Brazilian territory, the rest were used to minimize errors edge generated during the interpolation process. The data were processed through the interpolation tools located in the Spatial Analyst Tools and Geostatistical Wizard of the ARCGIS 10 program. For model validation a resampling technique was used, denominated Jackknife, which was of prime importance to standardize the minimum and maximum calculated values and corroborating with the waste generated attainment in the process. To compare the models six statistical criteria were used, in which Kriging , Topo to Raster and Global Polynomial Interpolation models showed the best results. The models obtained by Spline Radial Basis Function methods did not honour the input data. However, IDW, Natural Neighbor and Global Polynomial Interpolation models generated inconsistent anomalies with the physical characteristics of the crust, while Local Polynomial Interpolation reproduced a smooth model suppressing important information about the crust. Among the analysed models, kriging and Top Raster were the best ones . With the publication of new data of crustal thickness for Brazil, models can be generated in a quickly, more practical and efficient way.

No description available.

<p>Water contamination is one of the major environmental issues that negatively impacts water quality of watersheds. It negatively affects drinking water and aquatic wildlife, which can indirectly have negative effects on everyone's health. Many different institutions collected samples of water from four of Canada's major watersheds and counted the number of bacteria in each sample. The data used in this paper was taken from one of these institutions and was analysed to investigate if agricultural waste impacts the water quality of these four watersheds. It was found that the agricultural waste produced from nearby farms significantly impacts the water quality of three of these watersheds. Principal component analysis was also done on these data, and it was found that all of the data can be expressed in terms of one variable without losing very much information of the data. The bootstrap distributions of the principal component analysis parameters were estimated, and it was found that the sampling distributions of these parameters are stable. There was also evidence that the variables in the data are not normally distributed and not all the variables are independent.</p> / Master of Science (MSc)

Submitted by Valquíria Barbieri (kikibarbi@hotmail.com) on 2018-05-04T22:14:33Z No. of bitstreams: 1 DISS_2014_Evaldo Oestreich Filho.pdf: 2713813 bytes, checksum: bf84148c7524af203433132e7a61a4b8 (MD5) / Approved for entry into archive by Jordan (jordanbiblio@gmail.com) on 2018-05-16T16:43:42Z (GMT) No. of bitstreams: 1 DISS_2014_Evaldo Oestreich Filho.pdf: 2713813 bytes, checksum: bf84148c7524af203433132e7a61a4b8 (MD5) / Made available in DSpace on 2018-05-16T16:43:42Z (GMT). No. of bitstreams: 1 DISS_2014_Evaldo Oestreich Filho.pdf: 2713813 bytes, checksum: bf84148c7524af203433132e7a61a4b8 (MD5) Previous issue date: 2014-02-20 / CAPES / O estudo foi desenvolvido no estado de Mato Grosso, em dois locais, sendo o primeiro localizado no município de Cuiabá, Mato Grosso, Brasil, denominada Área 1 e o segundo localizado no município de Chapada dos Guimarães denominado Área 2. O presente trabalho teve como objetivos caracterizar o estrato arbóreo das comunidades de cerrado estudadas; quanto a riqueza, a florística, a estrutura fitossociológica e diversidade; determinar a similaridade entre cerrados fragmentos estudados e compará-los quanto à florística, estrutura fitossociológica e diversidade. Para este estudo foram alocadas 30 parcelas na área 1 e 32 parcelas na área 2, todas com tamanhos 20 m x 20 m (400 m²), totalizando 12.000 m² e 12.800 m², respectivamente nas Áreas 1 e 2. A suficiência de amostragem foi obtida com base na análise da curva do coletor por meio do estimador não-paramétrico Jackknife, a diversidade através do índice de Shannon, a equabilidade através do índice de Simpson e a similaridade das áreas foi calculada por meio do índice de Sorensen. As espécies foram organizadas de acordo com as famílias reconhecidas pelo Angiosperm Phylogeny Group III. Foram encontradas nas áreas 1 e 2, um total de 3139 indivíduos arbóreos distribuídos em 82 espécies entre indivíduos com DAB ≥ 5 cm, sendo 8 espécies identificadas somente em nível de gênero. Essas espécies distribuem-se entre 57 gêneros, 31 famílias botânicas. As famílias mais bem representadas em número de espécies foram: Fabaceae, Myrtaceae, Vochysiaceae, Apocynaceae, Annonaceae e Bignoniaceae. As espécies Qualea parviflora e Kielmeyera coriacea estiveram presentes em todas as parcelas levantadas no fragmento de cerrado estudado no município de Cuiabá, e, portanto, obtiveram frequência absoluta de 100%. As espécies Couepia grandiflora, Davilla elliptica e Qualea grandiflora estiveram presentes em 93,75% de todas as parcelas levantadas na área pertencente à Chapada dos Guimarães. A diversidade e a equabilidade, respectivamente, pelo índice de Shannon e de Simpson, foram: 3,34 e 0,9520 na área 1; 3,56 e 0,9559 na área 2. A similaridade entre as áreas estudadas foi de 42% pelo índice de Sorensen. Os resultados encontrados confirmaram que as famílias Fabaceae, Myrtaceae e Vochysiaceae são as mais representativas floristicamente nos cerrados do Brasil Central. / The study was conducted in the state of Mato Grosso, in two places, the first located in the the municipality of Cuiaba , Mato Grosso, Brazil and called Area 1 and the second located in the municipality of Chapada dos Guimarães called Area 2. The objective of this study was to characterize the arboreal stratum of the communities of cerrado, as richness, floristic, phytosociological structure and diversity; determine the similarity between cerrado fragments studied and compare them with respect to floristic, phytosociological structure and diversity. For this study, 30 plots were allocated in Area 1 and 32 plots in Area 2, all of them with size 20 m x 20 m (400 m²), with total 12,000 m² and 12,800 m², respectively in areas 1 and 2. The sampling adequacy was obtained based on the analysis of the curve through nonparametric Jackknife estimator, the diversity was calculated using the Shannon index, the equability across the Simpson index and the similarity of the areas was calculated using the Sorensen index. The species were arranged according to families recognized by the Angiosperm Phylogeny Group III. Were found a total of 3139 individual trees belonging to 82 species among individuals with the diameter average the base (DAB) ≥ 5 cm, with only 8 species identified at genus. These species is distributed among 57 genera, 31 botanical families. The families that had the best representations in number of species were: Fabaceae, Myrtaceae, Vochysiaceae, Apocynaceae, Annonaceae and Bignoniaceae. The species Qualea parviflora and Kielmeyera coriacea were present in all plots raised in cerrado fragment in Cuiabá, so had absolute frequency of 100%. The species Couepia grandiflora, Davilla elliptica and Qualea grandiflora were present in 93.75% of all parcels raised in Chapada of Guimarães. The diversity and the equability, respectively, by Shannon and Simpson indexes, were: 3.34 and 0.9520 in Area 1; 3.56 and 0.9559 in Area 2. The similarity between study areas showed 42% similarity by Sorensen index. The results confirmed that the Fabaceae, Myrtaceae and Vochysiaceae families are the most representative flora in cerrado of central Brazil.

A dissertação apresenta os resultados obtidos com o desenvolvimento de um programa de computação abrangente em interface gráfica para ambiente Windows visando a estimativa de parâmetros de interesse genético-populacional (freqüências alélicas, respectivos erros-padrão e intervalos de confiança a 95%) e o teste de hipóteses genéticas (equilíbrio de Hardy-Weinberg e análise de estruturação hierárquica populacional), por meio de métodos tradicionais e por meio de testes exatos obtidos com procedimentos de simulação (bootstrap e jackknife). / The dissertation presents the results obtained with the development of a comprehensive computation program (software), running on the Windows (MS) graphic interface, with the aim of: (a) estimating parameters of population genetic interest (such as allelic frequencies and their corresponding standard errors and 95% confidence intervals); and (b) performing the testing of genetic hypotheses (Hardy-Weinberg population ratios and analysis of population hierarchical structure) by means of traditional methods as well as through exact tests obtained with computer simulation procedures (bootstrap and jackknife methods).

A dissertação apresenta os resultados obtidos com o desenvolvimento de um programa de computação abrangente em interface gráfica para ambiente Windows visando a estimativa de parâmetros de interesse genético-populacional (freqüências alélicas, respectivos erros-padrão e intervalos de confiança a 95%) e o teste de hipóteses genéticas (equilíbrio de Hardy-Weinberg e análise de estruturação hierárquica populacional), por meio de métodos tradicionais e por meio de testes exatos obtidos com procedimentos de simulação (bootstrap e jackknife). / The dissertation presents the results obtained with the development of a comprehensive computation program (software), running on the Windows (MS) graphic interface, with the aim of: (a) estimating parameters of population genetic interest (such as allelic frequencies and their corresponding standard errors and 95% confidence intervals); and (b) performing the testing of genetic hypotheses (Hardy-Weinberg population ratios and analysis of population hierarchical structure) by means of traditional methods as well as through exact tests obtained with computer simulation procedures (bootstrap and jackknife methods).

In this dissertation, we first develop new statistical methodologies for small area estimation problems with measurement errors. The prediction of small area means for the unit-level regression model with the functional measurement error in the area-specific covariate is considered. We obtain the James-Stein (JS) estimate of the true area-specific covariate. Consequently, we construct the pseudo Bayes (PB) and pseudo empirical Bayes (PEB) predictors of small area means and estimate the mean squared prediction error (MSPE) associated with each predictor. Secondly, we modify the point estimation of the true area-specific covariate obtained earlier such that the histogram of the predictors of the small area means gets closer to its true one. We propose the constrained Bayes (CB) estimate of the true area-specific covariate. We show the superiority of the CB over the maximum likelihood (ML) estimate in terms of the Bayes risk. We also show the PB predictor of the small area mean based on the CB estimate of the true area-specific covariate dominates its counterpart based on the ML estimate in terms of the Bayes risk. We compare the performance of different predictors of the small area means using measures such as sensitivity, specificity, positive predictive value, negative predictive value, and MSPE. We believe that using the PEB and pseudo hierarchical Bayes predictors of small area means based on the constrained empirical Bayes (CEB) and constrained hierarchical Bayes (CHB) offers higher precision in recognizing socio-economic groups which are in danger of the prehypertension. Clustering the small areas to understand the behavior of the random effects better and accordingly, to predict the small area means is the final problem we address. We consider the Fay-Herriot model for this problem. We design a statistical test to evaluate the assumption of the equality of the variance components in different clusters. In the case of rejection of the null hypothesis of the equality of the variance components, we implement a modified version of Tukey's method. We calculate the MSPE to evaluate the effect of the clustering on the precision of predictors of the small area means. We apply our methodologies to real data sets. / February 2017