abstract: Urban growth, from regional sprawl to global urbanization, is the most rapid, drastic, and irreversible form of human modification to the natural environment. Extensive land cover modifications during urban growth have altered the local energy balance, causing the city warmer than its surrounding rural environment, a phenomenon known as an urban heat island (UHI). How are the seasonal and diurnal surface temperatures related to the land surface characteristics, and what land cover types and/or patterns are desirable for ameliorating climate in a fast growing desert city? This dissertation scrutinizes these questions and seeks to address them using a combination of satellite remote sensing, geographical information science, and spatial statistical modeling techniques. This dissertation includes two main parts. The first part proposes to employ the continuous, pixel-based landscape gradient models in comparison to the discrete, patch-based mosaic models and evaluates model efficiency in two empirical contexts: urban landscape pattern mapping and land cover dynamics monitoring. The second part formalizes a novel statistical model called spatially filtered ridge regression (SFRR) that ensures accurate and stable statistical estimation despite the existence of multicollinearity and the inherent spatial effect. Results highlight the strong potential of local indicators of spatial dependence in landscape pattern mapping across various geographical scales. This is based on evidence from a sequence of exploratory comparative analyses and a time series study of land cover dynamics over Phoenix, AZ. The newly proposed SFRR method is capable of producing reliable estimates when analyzing statistical relationships involving geographic data and highly correlated predictor variables. An empirical application of the SFRR over Phoenix suggests that urban cooling can be achieved not only by altering the land cover abundance, but also by optimizing the spatial arrangements of urban land cover features. Considering the limited water supply, rapid urban expansion, and the continuously warming climate, judicious design and planning of urban land cover features is of increasing importance for conserving resources and enhancing quality of life. / Dissertation/Thesis / Doctoral Dissertation Geography 2016

This thesis explores some of the limitations and implications of using multiple regression analysis in transportation models. Specifically it investigates how the problem of multicollinearity, which results from using intercorrelated variables in trip generation models, adversely affects the validation of hypotheses, discovery of underlying relationships and prediction. The research methodology consists of a review of the literature on trip generation analysis and a theoretical exposition on multicollinearity. Secondly, trip generation data for Greater Vancouver (1968) is used for empirical analysis. Factor analysis and multiple regression techniques are employed. The results demonstrate that multicollinearity is both an explanatory and prediction problem which can be overcome by a combined factor analytic and regression method. This method is also capable of identifying and incorporating causal relationships between land use and trip generation into a single model. It is concluded that the distinction between the explanatory, analytic and predictive abilities of a regression model is artificial, and that greater emphasis on theorizing in model-construction is needed. . / Applied Science, Faculty of / Community and Regional Planning (SCARP), School of / Graduate

No description available.

Doctor of Philosophy / Department of Economics / Lance J. Bachmeier / This dissertation consists of three essays in nonlinear macroeconomic modeling and econometrics. In the first essay, we decompose oil price movements into oil demand (stock market) shocks and oil supply (oil-market) shocks, and examine the response of the stock market to these shocks. We find that when oil prices are “net-increasing”, a stock market shock that causes the S&P 500 to rise by one percentage point will cause the price of oil to rise approximately 0.2 percentage points, with a statistically significant positive effect one day after the stock market shock. On the other hand, the response of the stock market to an oil market shock is a decline of 6.8 percent when the price of oil doubles. For other days, the initial response of the oil market to a stock market shock is the same as in the net oil price increase case (by construction). We then analyze the response of monetary policy to the identified stock market and oil market shocks and find that short-term interest rates respond to the stock market shocks but not the oil market shocks. Finally, we evaluate the predictive power of the decomposed stock market and oil shocks relative to the change in the price of oil. We find statistically significant gains in both the in-sample fit and out-of-sample forecast accuracy when using the identified stock market and oil market shocks rather than the change in the price of oil. The second essay revisits the statistical specification of near-multicollinearity in the logistic regression model using the Probabilistic Reduction approach. We argue that the ceteris paribus clause invoked with near-multicollinearity is rather misleading. This assumption states that one can assess the impact of near-multicollinearity by holding the parameters of the logistic regression model constant, while examining the impact on their standard errors and t-ratios as the correlation (\rho) between the regressors increases. Using the Probabilistic Reduction approach, we derive the parameters (and related statisitics) of the logistic regression model and show that they are functions of \rho , indicating the ceteris paribus clause in the traditional account of near multicollinearity is unattainable. Monte carlo simulations in the paper confirm these findings. We also show that traditional near-multicollinearity diagnostics, such as the variance inflation factor and condition number can fail to detect near-multicollinearity. Overall, the paper finds that near-multicollinearity in the logistic model is highly variable and may not lead to the problems indicated by the traditional account. Therefore, unexpected, unreliable or unstable estimates and inferences should not be blamed on near-multicollinearity. Rather the modeler should return to economic theory or statistical respecification of their model to address these problems. The third essay examines the correlations between income inequality and economic growth using a panel of income distribution data for 3,109 counties of the U.S. We examine the non-spatial dynamic correlations between county inequality and growth using a System GMM approach, and find significant negative relationships between changes in inequality in one period and growth in the subsequent period. We show that this finding is robust across different sample sizes. We further argue that because the space-specific time-invariant variables that affect economic growth and inequality can differ significantly across counties, failure to incorporate spatial effects into a model of growth and inequality may lead to biased results.We assume that dependence among counties only arises from the disturbance process, hence the estimation of a spatial error model. Our results indicate that the bias in the parameter for inequality amounts to about 2.66 percent, while that for initial income amounts to about 21.51 percent.

Neste trabalho estudamos os efeitos da multicolinearidade em modelos de regressão logística e apresentamos estimadores viesados para que tais efeitos fossem minimizados. Primeiramente, o modelo de regressão logística e o processo para a estimação dos parâmetros foram apresentados. Foram feitos, também, alguns testes para avaliar a significância dos mesmos, bem como técnicas para analisar a qualidade do ajuste do modelo. Em seguida, os efeitos da multicolinearidade na estimação dos parâmetros e na sua inferência foram avaliados, bem como técnicas para o seu diagnóstico. Para amenizar o efeito deste problema, apresentamos dois estimadores alternativos ao de máxima verossimilhança: estimador em cristas e estimador em componentes principais. Comparamos, então, o desempenho dos três estimadores na forma de um estudo de simulação e de uma aplicação em um conjunto de dados reais. O principal resultado obtido foi que, na presença de multicolinearidade, os estimadores alternativos conseguiram um melhor ajuste em comparação ao de máxima verossimilhança, além de minimizar os seus efeitos. / This work proposes the use of some biased estimators to investigate whether is possible minimize the multicollinearity effects in logistic regression models. Initially, the latter model was presented, as well as its fitting process (therefore obtaining the maximum likelihood estimator), some tests to evaluate the significance of the parameters and techniques to analyze goodness of fit were also considered. Furthermore, the effects of multicollinearity in the fitting process and in the parameters inference were discussed, as well as techniques to identify the presence of multicollinearity. In order to diminish the effect of this problem, two alternative estimators were presented: ridge estimator and principal component estimator. Therefore, these three estimators performances were compared using a simulation study and applied in a real data set. The manly conclusion was that, in the presence of multicollinearity, the alternative estimators performed better than the maximum likelihood estimator, besides reducing its effects.

Neste trabalho estudamos os efeitos da multicolinearidade em modelos de regressão logística e apresentamos estimadores viesados para que tais efeitos fossem minimizados. Primeiramente, o modelo de regressão logística e o processo para a estimação dos parâmetros foram apresentados. Foram feitos, também, alguns testes para avaliar a significância dos mesmos, bem como técnicas para analisar a qualidade do ajuste do modelo. Em seguida, os efeitos da multicolinearidade na estimação dos parâmetros e na sua inferência foram avaliados, bem como técnicas para o seu diagnóstico. Para amenizar o efeito deste problema, apresentamos dois estimadores alternativos ao de máxima verossimilhança: estimador em cristas e estimador em componentes principais. Comparamos, então, o desempenho dos três estimadores na forma de um estudo de simulação e de uma aplicação em um conjunto de dados reais. O principal resultado obtido foi que, na presença de multicolinearidade, os estimadores alternativos conseguiram um melhor ajuste em comparação ao de máxima verossimilhança, além de minimizar os seus efeitos. / This work proposes the use of some biased estimators to investigate whether is possible minimize the multicollinearity effects in logistic regression models. Initially, the latter model was presented, as well as its fitting process (therefore obtaining the maximum likelihood estimator), some tests to evaluate the significance of the parameters and techniques to analyze goodness of fit were also considered. Furthermore, the effects of multicollinearity in the fitting process and in the parameters inference were discussed, as well as techniques to identify the presence of multicollinearity. In order to diminish the effect of this problem, two alternative estimators were presented: ridge estimator and principal component estimator. Therefore, these three estimators performances were compared using a simulation study and applied in a real data set. The manly conclusion was that, in the presence of multicollinearity, the alternative estimators performed better than the maximum likelihood estimator, besides reducing its effects.

<p>This project &quot / An application of multiple regression in exchange rate arrangement&quot / focused on the processes followed by different countries when choosing an exchange rate regime for currency stabilization. It analyses the consequences faced by emerging markets as a result of changes in volatility of developed countries&rsquo / currencies (American Dollar, Japanese Yen, EURO, British Pound and the Canadian Dollar).</p>

<p>This project &quot / An application of multiple regression in exchange rate arrangement&quot / focused on the processes followed by different countries when choosing an exchange rate regime for currency stabilization. It analyses the consequences faced by emerging markets as a result of changes in volatility of developed countries&rsquo / currencies (American Dollar, Japanese Yen, EURO, British Pound and the Canadian Dollar).</p>

Forage crops practice an important role in animal production in southern Brazil. Among the species, oat stands out due to its higher crop area in winter, occupying an area of 3.8 million hectares in the state of Rio Grande do Sul. For their proper planting and establishment, seeds are produced which must contain high standards of quality, which is measured by analytical laboratories. Given this, the present study aimed to evaluate the physical and physiological attributes of oat seeds, associating quality seeds to the production profile and the possible effects caused by meteorological factors. Was also designed to identify the variables that correlate with the percentage of pure seed and seedling emergence, identify the presence of multicollinearity, the most important variables in relation to the main dependent variable, percentage of normal seedlings, and group the samples for their degrees of similarity. 2,910 samples were assessed, 2,229 seed analysis derived from the seed production process, 357 analyzes for own use of seeds and 324 analyzes tetrazolium analyzed by the laboratory of the UNIJUÍ Agronomy Course seeds, following the methodology described in the rules seed analysis. The results were submitted to analysis of descriptive statistics, the dispersion of the data, the Pearson linear correlation coefficients were estimated, the diagnosis of multicollinearity, the direct and indirect effects through path analysis and grouping between samples. The seeds produced according to the national seed and seedling system showed excellent levels of physical and physiological quality in the years 2006 to 2010. Between 2011- 2014, 14 and 14.5% of the seeds have been compromised by the presence of other seed species cultivated and tolerated harmful, respectively. The proper use of seeds showed wide variability with 18.1 and 31.7% of samples below the standard for germination in the years 2006 to 2010 and 2011-2014, respectively, while the samples analyzed by the tetrazolium test showed disapproval levels 19.4 and 12.5%, respectively. It is noteworthy that the seed quality is related to the years with levels of rainfall and temperatures appropriate to the vegetative development, physiological maturity and harvest. The variable normal seedlings showed the highest correlation, a negative sign with dead seeds. Abnormal seedlings variables and dead seeds showed the highest direct effects on germination percentage, negative sign and cluster analysis revealed the existence of three similarity groups of seeds produced according to the national system of seed plants and four groups in seed own use. / As plantas forrageiras desempenham importante papel na produção animal na Região Sul do Brasil. Dentre as espécies, a aveia preta se destaca por apresentar maior área de cultivo no inverno, ocupando uma área de 3,8 milhões de hectares no Estado do Rio Grande do Sul. Para a sua adequada semeadura e estabelecimento, são produzidas sementes que devem conter elevados padrões de qualidade, que é aferida por laboratórios de análise. Frente a isso, a presente pesquisa teve como objetivos avaliar os atributos físicos e fisiológicos de sementes de aveia preta, associar a qualidade de sementes ao perfil de produção e aos possíveis efeitos provocados por fatores meteorológicos. Visou, ainda, identificar as variáveis que se correlacionam com a porcentagem de sementes puras e a emergência de plântulas, identificar a presença de multicolinearidade, as variáveis mais importantes em relação à variável dependente principal, porcentagem de plântulas normais, e agrupar a amostras por seus graus de parecença. Foram avaliadas 2.910 amostras, sendo 2.229 análises de sementes oriundas do processo de produção de sementes, 357 análises de sementes de uso próprio e 324 análises de tetrazólio analisadas pelo laboratório de análise de sementes do Curso de Agronomia da UNIJUÍ, seguindo a metodologia descrita nas regras de análise de sementes. Os resultados foram submetidos às análises de estatísticas descritivas, à dispersão dos dados, foram estimados os coeficientes de correlação linear de Pearson, o diagnóstico de multicolinearidade, os efeitos diretos e indiretos através da análise de trilha e o agrupamento entre as amostras. As sementes produzidas segundo o sistema nacional de sementes e mudas apresentaram excelentes níveis de qualidade física e fisiológica nos anos de 2006 a 2010. Entre os anos de 2011 a 2014, 14 e 14,5% das sementes foram comprometidas pelas presenças de outras sementes de espécies cultivadas e de nocivas toleradas, respectivamente. As sementes de uso próprio apresentaram ampla variabilidade com 18,1 e 31,7% de amostras abaixo dos padrão para germinação nos anos de 2006 a 2010 e 2011 a 2014, respectivamente, enquanto que as amostras analisadas pelo teste de tetrazólio apresentaram níveis de reprovação de 19,4 e 12,5 %, respectivamente. Destaca-se que a qualidade fisiológica das sementes está relacionada aos anos com níveis de precipitações e temperaturas adequadas ao desenvolvimento vegetativo, maturidade fisiológica e colheita. A variável plântulas normais apresentou maior correlação, de sinal negativo, com sementes mortas. As variáveis plântulas anormais e sementes mortas apresentaram os maiores efeitos diretos sobre porcentagem de germinação, de sinal negativo e a análise de agrupamento revelou a existência de três grupos de parecença em sementes produzidas segundo o sistema nacional de sementes mudas e de quatro grupos em sementes de uso próprio.

Magister Scientiae - MSc / This project "An application of multiple regression in exchange rate arrangement" focused on the processes followed by different countries when choosing an exchange rate regime for currency stabilization. It analyses the consequences faced by emerging markets as a result of changes in volatility of developed countries&rsquo; currencies (American Dollar, Japanese Yen, EURO, British Pound and the Canadian Dollar). / South Africa