No description available.

No description available.

No description available.

No description available.

Most research on confirmatory factor analysis using global fit indices (AIC, BIC, AICc, and CAIC) has been in the structural equation modeling framework. Little research has been done concerning application of these indices to item response models, especially within the framework of multidimensional Rasch analysis. The results of two simulations studies that investigated how sample size, between-dimension correlation, and test length affect the accuracy of these indices in model recovery using a multidimensional Rasch analysis are described in this dissertation. The first study analyzed dichotomous data, with model-to-data misfit as an additional independent variable. The second study analyzed polytomous data, with rating scale structure as an additional independent variable. The interaction effect between global fit index and between-dimension correlation had very large effect sizes in both studies. At higher values of between-dimension correlation, AIC indicated the correct two-dimension generating structure slightly more often than does the BIC or CAIC. The correlation by test length interaction had an odds ratio indicating practical importance in the polytomous study but not the dichotomous study. The combination of shorter tests and higher correlations resulted in a difficult-to-detect distinction being modeled with less statistical information. The correlation by index interaction in the dichotomous study had an odds ratio indicating practical importance. As expected, the results demonstrated that violations of the Rasch model assumptions are magnified at higher between-dimension correlations. Recommendations for practitioners working with highly correlated multidimensional data include creating moderate length (roughly 40 items) instruments, minimizing data-to-model misfit in the choice of model used for confirmatory factor analysis (MRCMLM or other MIRT models), and making decisions based on multiple global indices instead of depending on one index in particular. / Ph. D.

In this dissertation, we develop a model selection and estimation methodology for nonstationary spatial fields. Large, spatially correlated data often cover a vast geographical area. However, local spatial regions may have different mean and covariance structures. Our methodology accomplishes three goals: (1) cluster locations into small regions with distinct, stationary models, (2) perform Bayesian model selection within each cluster, and (3) correlate the model selection and estimation in nearby clusters. We utilize the Conditional Autoregressive (CAR) model and Ising distribution to provide intra-cluster correlation on the linear effects and model inclusion indicators, while modeling inter-cluster correlation with separate Gaussian processes. We apply our model selection methodology to a dataset involving the prediction of Brook trout presence in subwatersheds across Pennsylvania. We find that our methodology outperforms the stationary spatial model and that different regions in Pennsylvania are governed by separate Gaussian process regression models. / Ph. D.

Doctor of Philosophy / Department of Entomology / James F. Campbell / Thomas W. Phillips / Movement is a fundamental feature of animals that impacts processes across multiple scales in space and time. Due to the heterogeneous and fragmented nature of habitats that make up landscapes, movement is not expected to be random in all instances, and an increase in fitness is an expected consequence for those that can optimize movement to find valuable and scarce recourses. I studied the movement of Rhyzopertha dominica (Coleoptera: Bostrichidae), one of the most important pests of stored grain worldwide, within and between resource patches. At a fine spatial scale, I identified factors that contribute to overall and upward movement in the grain mass. Three-week-old insects tented to stay closer to the surface than one or two-week-old insects. Females tended to be more active and to explore more than males. I also found that males tended to stay closer to the surface than females and that might be related to the ability to attract females from outside the patch since there was no significant difference regarding female’s attraction within the grain patch. Interaction with feeding sites or other individuals of the same sex creates positive feedback and a more clumped spatial pattern of feeding and foraging behavior. On the other hand, interaction with individuals of different sex creates negative feedback and a more random or overdispersed pattern. At a broad spatial scale, I studied the long-term consequence of R. dominica movement on the development of population structure within the U.S. To evaluate population structure, I used reduced representation of the genome followed by direct sequencing of beetles collected from different locations across the U.S where wheat or rice is produced and stored. Ecoregions were more important in explaining structure of R. dominica populations than crop type. I also found significant isolation by distance; however, model selection primarily elected grain production and movement variables to explain population differentiation and diversity. Understanding animal movement is essential to establishing relationships between distribution and surrounding landscape, and this knowledge can improve conservation and management strategies.

Potentially valuable information about the underlying data generating process of a dependent variable is often lost when an independent variable is transformed to fit into the same sampling frequency as a dependent variable. With the mixed data sampling (MIDAS) technique and increasingly available data at high frequencies, the issue of choosing an optimal sampling frequency becomes apparent. We use financial data and the MIDAS technique to estimate thousands of regressions and forecasts in the quarterly, monthly, weekly, and daily sampling frequencies. Model fit and forecast performance measurements are calculated from each estimation and used to generate summary statistics for each sampling frequency so that comparisons can be made between frequencies. Our regression models contain an autoregressive component and five additional independent variables and are estimated with varying lag length specifications that incrementally increase up to five years of lags. Each regression is used to forecast a rolling, one and two-step ahead, static forecast of the quarterly Yen and U.S Dollar spot exchange rate. Our results suggest that it may be favourable to include high frequency variables for closer modeling of the underlying data generating process but not necessarily for increased forecasting performance. / Graduate / 0501 / 0508 / 0511 / benjamincann@gmail.com

A conversão do uso do solo é uma das principais ameaçadas à fauna, pois a paisagem torna-se fragmentada e as áreas ocupadas por vegetação nativa são reduzidas, podendo alterar a estrutura das comunidades animais. Neste estudo, a comunidade de mamíferos de médio e grande porte (> 1 kg) foi avaliada em uma paisagem fragmentada com matriz de eucalipto, no sul do Estado de São Paulo, municípios de Buri e Capão Bonito (23°52\'47\" S; 48°23\'24\" O), através de três métodos distintos instalados conjuntamente (parcelas de areia, camera-trap e scent stations - nesta última foram utilizadas iscas atrativas específicas para carnívoros e onívoros, Canine Call® e Pro\'s Choice®). Os objetivos desta dissertação foram: i) avaliar a estrutura dessa comunidade em função da estrutura da paisagem; ii) comparar métodos de amostragem utilizados, discutindo sua aplicação. O levantamento das espécies foi realizado em cinco campanhas de cinco dias cada, entre 2010-2012, com unidades amostrais instaladas na matriz, nos corredores e nos fragmentos florestais, sendo que os elementos da paisagem foram avaliados em buffers de 250, 500, 1000 e 2000 m. Foram registradas 20 espécies de mamíferos de médio e grande porte, sendo a maioria (n=18) encontrada nos ambientes ocupados por vegetação nativa. Embora em menor número, várias espécies foram registradas na matriz. Outras medidas de biodiversidade como riqueza de grupos e diversidade funcional foram avaliados, no entanto, foram redundantes ao número de espécies. As proporções de vegetação nativa, a 250 e 2000 m, foram as mais importantes para explicar a frequência de registros de diversas espécies. Em termos de composição, os conjuntos de espécies obtidos nos remanescentes e corredores florestais foram semelhantes. Dentre os métodos de amostragem, as parcelas de areia demonstraram ser mais eficientes em estudos curtos, porém, seu custo financeiro é significativamente superior para estudos de longo prazo. A utilização de iscas odoríferas é uma inovação na região neotropical, apesar de suas incertezas na atração da fauna. Uma desvantagem desses métodos é que várias espécies não podem ser precisamente identificadas, dada a semelhança dos rastros, o que raramente ocorre com a utilização de cameras-trap, onde os espécimes são visualizados em fotografias, permitindo uma identificação segura. As diferentes metodologias utilizadas apresentaram taxas de registros distintas, ou seja, não foram capazes de identificar a mesma comunidade, entretanto, as estimativas de riqueza, quando avaliadas separadamente, indicam que os três atingiriam o mesmo resultado final. Este estudo sugere que paisagens silviculturais fragmentadas podem ser importantes para a conservação de mamíferos se bem planejadas, como a manutenção de manchas de habitat em bons estados de conservação e corredores que conectem os elementos florestais. Além disso, a utilização dos métodos de amostragem deve ser planejados de acordo com a finalidade do estudo. / The land use conversion is a major threat to wildlife because the landscape becomes fragmented and the areas occupied by native vegetation are reduced, altering the structure of animal communities. In this study, the medium to large-sized mammal community (> 1 kg) was evaluated in a fragmented landscape in eucalyptus matrix, in the southern São Paulo State, Buri and Capão Bonito municipalities (23°52\'47\" S, 48°23\'24\" W), using three different methods installed together (sand plots, camera-trap and scent stations, where were used specific baits to carnivores and omnivores, Canine Call® and Pro\'s Choice®). The objectives of this work were: i) to evaluate the structure of the community as a function of landscape structure, ii) compare sampling methods used, discussing their application. The species sampling was conducted in five campaigns of five days each, between 2010-2012, with sampling units installed in the matrix, corridors and forest fragments. The landscape elements were evaluated in buffers with 250, 500, 1000 and 2000 m around. We recorded 20 medium and large mammals species, the majority (n=18) found in y native vegetation (corridors and habitat patches). Although only several species were recorded in the matrix. Other measures of biodiversity as a functional group richness and functional diversity were evaluated, however, the results were the same obtained to species richness. The proportions of native vegetation, 250 and 2000 m, were most important in explaining the records frequency for many species. In terms of composition, the assemblage obtained in the habitat patches and forest corridors were similar. Among the methods of sampling, the sand plots was more effective in short-term assessment, however, its financial cost is significantly higher for long-term studies. The use of lure in scent stations is an innovation in the Neotropical, despite their uncertainties in attracting the animals. A disadvantage of these methods is that several species cannot be accurately identified because of the similarity of the tracks, which rarely occurs with the use of cameras-trap, where the specimens are shown in photographs, allowing a reliable identification. The different methodologies used showed different rates of records, or were not able to identify the same community, however, the richness estimates, when evaluated separately, indicate that the three would achieve the same result. This study suggests that forestry fragmented landscapes may be important for mammals conservation if well planned, such as the maintenance of habitat patches and corridors that connect the remaining habitat fragments. Furthermore, the use of sampling method must be planned according to the purpose of study.

A comparação e seleção de modelos estatísticos desempenham um papel fundamental dentro da análise econométrica. No que se trata especificamente da avaliação de modelos não aninhados, o procedimento de teste denominado de J-Teste aparece como uma ferramenta de uso freqüente nessa literatura. De acordo com apontamentos, entre os anos de 1984 e 2004 o J-Teste foi citado em 497 artigos pertinentes. Diferentemente do J-Teste, as abordagens Bayesianas possuem um potencial de aplicabilidade ainda pouco explorado na literatura, dado que são metodologicamente coerentes com os procedimentos inferenciais da econometria. Nesse sentido, o objetivo do presente trabalho é o de avaliar a aplicabilidade do procedimento de teste Bayesiano FBST para a comparação de modelos econométricos não aninhados. Implementando-se o FBST para os mesmos dados de estudos estatísticos relevantes na Teoria Econômica, tais como Bremmer (2003) (Curva de Phillips) e Caporale e Grier (2000) (determinação da taxa de juros real), constata-se que os resultados obtidos apontam para conclusões semelhantes daquelas delineadas com a utilização do J-Teste. Além disso, ao se utilizar a noção de função poder para avaliar ambos os procedimentos de teste, observa-se que sob certas condições as chances de erro expressas pelo Erro Tipo I e Erro Tipo II se tornam relativamente próximas. / The comparison and selection of statistical models play an important role in econometric analysis. Dealing with evaluation of non nested models, the test procedure called J-Test is a frequently used tool in the literature. Accordingly to statistics, between the years 1894 and 2004 the J-Test was cited on 497 pertinent articles. Differently from J-Test, the Bayesian theories have an unexplored applicability potential in the literature, once they are methodologically coherent with the standard procedures of inference in econometrics. In this sense, the objective of this essay is to evaluate the applicability of the Bayesian procedure FBST to comparison of non nested econometric models. Implementing the FBST to the same data of some relevant statistical studies in Economic Theory, like Bremmer (2003) (Phillips Curve) and Caporale and Grier (2000) (real interest rate determination), it can be seen that the results obtained point to the same conclusions as that attained with J-Test utilization. Besides that, when implementing the power function to evaluate both test procedures, it can be observed that under some conditions the error chances expressed by Error Type I and Error Type II become relatively close.