What is the relationship between gender inequality and resource wealth in the Middle East? Why has progression of women’s rights in the Middle East advanced at a comparatively slower rate than most of the world? Conventional wisdom attributes the continued significant gaps in gender equality to the region’s strong patriarchic culture associated with Islam. However, recent statistical analysis conducted by Michael Ross suggests a correlation between oil production and women’s rights. This thesis examines an emerging schism in the literature and evaluates the relationship between social and political emancipation of women relative to Islam and oil wealth. The findings of this examination conclude that Ross’s theoretical framework is incomplete: Islamic law is a key causal mechanism left out of his examination. Furthermore, Islamic law has a more comprehensive negative impact on women’s social and political rights than oil wealth.

Une approche fondamentale pour la résolution de problémes de contrôle optimal est basée sur le principe de programmation dynamique. Ce principe conduit aux équations d'Hamilton-Jacobi, qui peuvent être résolues numériquement par des méthodes classiques comme la méthode des différences finies, les méthodes semi-lagrangiennes, ou les schémas antidiffusifs. À cause de la discrétisation de l'espace d'état, la dimension des problèmes de contrôle pouvant être abordés par ces méthodes classiques est souvent limitée à 3 ou 4. Ce phénomène est appellé malédiction de la dimension. Cette thèse porte sur les méthodes numériques max-plus en contôle optimal deterministe et ses analyses de convergence. Nous étudions et developpons des méthodes numériques destinées à attenuer la malédiction de la dimension, pour lesquelles nous obtenons des estimations théoriques de complexité. Les preuves reposent sur des résultats de théorie de Perron-Frobenius non linéaire. En particulier, nous étudions les propriétés de contraction des opérateurs monotones et non expansifs, pour différentes métriques de Finsler sur un cône (métrique de Thompson, métrique projective d'Hilbert). Nous donnons par ailleurs une généralisation du "coefficient d'ergodicité de Dobrushin" à des opérateurs de Markov sur un cône général. Nous appliquons ces résultats aux systèmes de consensus ainsi qu'aux équations de Riccati généralisées apparaissant en contrôle stochastique.

In this dissertation we developed theoretical models to analyse Gaussian and multinomial distributions. The analysis is focused on classification in high dimensional feature spaces and provides a basis for dealing with issues such as data sparsity and feature selection (for Gaussian and multinomial distributions, two frequently used models for high dimensional applications). A Naïve Bayesian philosophy is followed to deal with issues associated with the curse of dimensionality. The core treatment on Gaussian and multinomial models consists of finding analytical expressions for classification error performances. Exact analytical expressions were found for calculating error rates of binary class systems with Gaussian features of arbitrary dimensionality and using any type of quadratic decision boundary (except for degenerate paraboloidal boundaries). Similarly, computationally inexpensive (and approximate) analytical error rate expressions were derived for classifiers with multinomial models. Additional issues with regards to the curse of dimensionality that are specific to multinomial models (feature sparsity) were dealt with and tested on a text-based language identification problem for all eleven official languages of South Africa. / Thesis (M.Ing. (Computer Engineering))--North-West University, Potchefstroom Campus, 2009.

In this dissertation we developed theoretical models to analyse Gaussian and multinomial distributions. The analysis is focused on classification in high dimensional feature spaces and provides a basis for dealing with issues such as data sparsity and feature selection (for Gaussian and multinomial distributions, two frequently used models for high dimensional applications). A Naïve Bayesian philosophy is followed to deal with issues associated with the curse of dimensionality. The core treatment on Gaussian and multinomial models consists of finding analytical expressions for classification error performances. Exact analytical expressions were found for calculating error rates of binary class systems with Gaussian features of arbitrary dimensionality and using any type of quadratic decision boundary (except for degenerate paraboloidal boundaries). Similarly, computationally inexpensive (and approximate) analytical error rate expressions were derived for classifiers with multinomial models. Additional issues with regards to the curse of dimensionality that are specific to multinomial models (feature sparsity) were dealt with and tested on a text-based language identification problem for all eleven official languages of South Africa. / Thesis (M.Ing. (Computer Engineering))--North-West University, Potchefstroom Campus, 2009.

This thesis studies the poor development performance of resource-rich developing economies, known as the resource curse. In the first chapter we provide a comprehensive literature review of the topic and the channels through which resource abundance can result in the resource curse. Issues of corruption and governance have been emphasised to be the main driver of the resource curse. This has been illustrated by a negative relationship between resource abundance and corruption control in the literature. However, there is a gap in how natural resources facilitate corruption. In the second chapter, using empirical analysis, we study the role of taxation in the relationship between natural resources and corruption. Taxation is usually seen as a social contract between citizens and government -- people pay taxes and in return they hold their government accountable for efficient allocation of their taxes. Resource abundance shifts the reliance of government from tax incomes to resource rents. People therefore, have no sustainable mechanism to hold their government responsible for corruption and wrongdoings inside public institutions. Using different econometric methods, Pooled OLS, Fixed Effects and 3SLS, our results show that natural resource revenues crowd out incomes from tax revenues. Meanwhile, taxation has a positive and significant impact on the control of corruption throughout our analysis. The results suggest that resource-rich developing countries should invest in building their tax systems to increase their non-resource tax revenues. This will increase state capacity and demand for accountability in the public sector among citizens and hence decreases corruption. Related to our second chapter, in the third chapter we study a cash transfer programme, known as oil-to-cash, which has gained support as a tool to re-establish taxation and fight corruption. Under such a plan, resource revenues are distributed directly among the public and then each citizen is taxed optimally. Through this, government relies directly and fully on its citizens for its income. Hence, taxation is reinstated and the social contract is revived. Within a general equilibrium model we show how this happens and what the implications are of the oil-to-cash programme for economic growth. Our results clearly show how corruption results in a resource curse. Furthermore, the model explains the variation that is seen in the degree of the resource curse across countries. The study also analyses the practical barriers of the oil-to-cash plan. The study suggests that parallel to (or even prior to) such a plan countries need to invest in building their tax system and increasing their administrative capacities.

We discuss efficient Bayesian estimation of dynamic covariance matrices in multivariate time series through a factor stochastic volatility model. In particular, we propose two interweaving strategies (Yu and Meng, Journal of Computational and Graphical Statistics, 20(3), 531-570, 2011) to substantially accelerate convergence and mixing of standard MCMC approaches. Similar to marginal data augmentation techniques, the proposed acceleration procedures exploit non-identifiability issues which frequently arise in factor models. Our new interweaving strategies are easy to implement and come at almost no extra computational cost; nevertheless, they can boost estimation efficiency by several orders of magnitude as is shown in extensive simulation studies. To conclude, the application of our algorithm to a 26-dimensional exchange rate data set illustrates the superior performance of the new approach for real-world data. / Series: Research Report Series / Department of Statistics and Mathematics

Orientador: Prof. Dr. Ronaldo Pratti / Dissertação (mestrado) - Universidade Federal do ABC, Programa de Pós-Graduação em Ciência da Computação, 2017. / O efeito de hubness, juntamente com a maldição de dimensionalidade, vem sendo estudado, sob diferentes oticas, nos ultimos anos. Os estudos apontam que este problema esta presente em varios conjuntos de dados do mundo real e que a presença de hubs (tendencia de alguns exemplos aparecem com frequencia na lista de vizinhos mais proximos de outros exemplos) traz uma serie de consequencias indesejaveis, como por exemplo, afetar o desempenho de classificadores. Em tarefas de mineração de texto, o problema depende tambem da maneira escolhida pra representar os documentos. Sendo assim o objetivo principal dessa dissertação é avaliar o impacto da formação de hubs em diferentes representações textuais. Ate onde vai o nosso conhecimento e durante o período desta pesquisa, não foi posivel encontrar na literatura um estudo aprofundado sobre as implicaçõess do efeito de hubness em diferentes representações textuais. Os resultados sugerem que as diferentes representações textuais implicam em corpus com propensão menor para a formação de hubs. Notou-se também que a incidencia de hubs nas diferentes representações textuais possuem in uencia similar em alguns classificadores. Analisamos tambem o desempenho dos classifcadores apos a remoção de documentos sinalizados como hubs em porçõess pre-estabelecidas do tamanho total do data set. Essa remoção trouxe, a alguns algoritmos, uma tendencia de melhoria de desempenho. Dessa maneira, apesar de nem sempre efetiva, a estrategia de identifcar e remover hubs com uma vizinhança majoritariamente ruim pode ser uma interessante tecnica de pre-processamento a ser considerada, com o intuito de melhorar o desempenho preditivo da tarefa de classificação. / The hubness phenomenon, associated to the curse of dimensionality, has been studied, from diferent perspectives, in recent years. These studies point out that the hubness problem is present in several real-world data sets and, as a consequence, the hubness implies a series of undesirable side efects, such as an increase in misclassifcation error in classification tasks. In text mining research, this problem also depends on the choice of text representation. Hence, the main objective of the dissertation is to evaluate the impact of the hubs presence in diferent textual representations. To the best of our knowledge, this is the first study that performs an in-depth analysis on the efects of the hub problem in diferent textual representations. The results suggest that diferent text representations implies in diferent bias towards hubs presence in diferent corpus. It was also noticed that the presence of hubs in dierent text representations has similar in uence for some classifiers. We also analyzed the performance of classifiers after removing documents agged as hubs in pre-established portions of the total data set size. This removal allows, to some algorithms, a trend of improvement in performance. Thus, although not always efective, the strategy of identifying and removing hubs with a majority of bad neighborhood may be an interesting preprocessing technique to be considered in order to improve the predictive performance of the text classification task.

The dissertation research will comprise three essays on the topic of the resource curse hypothesis and its mechanisms. The phenomenon of low economic growth in resource-rich regions is recognized as the “resource curse”. These essays will contribute to an understanding of the regional resource-growth relation within a nation. Essay one tests the resource curse hypothesis at the U.S. state level. With a system of equations model, I decompose the overall resource effect to account for the two leading explanations — crowding-out and institution effects, thus investigate whether the institutions mediate the crowding-out effects. I did not find evidence of an overall negative effect on growth by resource wealth. Both the crowding-out and institution appear present, but they offset: the resource boom crowds out industrial investments, but good institutions mitigate the overall effect. Resources do reduce growth in states with low-quality institutions, including Louisiana, Oklahoma, and Texas. Essay two compares the effects of resource revenues on the economic growth and growth-related factors across Chinese provinces and American states, using panel data from 1990 to 2015. With the Instrumental Variable (IV) strategy, I show that regions with higher resource revenues grow faster than other regions in both China and the U.S. The positive resource effect is larger and more statistically significant in the U.S. Further testing impacts of three resource-related policies in China, e.g. the market price reform, the fiscal reform, and the Western Development Strategy, I show that the market price reform together with the privatization process on coal resources contribute the positive resource effect in China. Though strong and positive resource – growth relations appear in both countries, evidence also suggests consistent negative resource effects on certain growth-related factors in both countries, such as educational attainments and R&D activities. Essay three explores the schooling response to the oil and gas boom, taking advantage of timing and spatial variation in oil and gas well drilling activities. Development of cost-reducing technologies at the time of higher crude oil and natural gas prices in the early 2000s has accelerated shale oil and gas extraction in the United States. I show that intensive drilling activities have decreased grade 11 and 12 enrollment over the 14 year study window − approximately 36 fewer students per county on average and overall, 41,760 fewer students across the 15 states enrolled considered in the analysis. On average, with one additional oil or gas well drilled per thousand initial laborers, grade 11 and 12 enrollment would decrease 0.24 percent at the county level, all else equal. I investigate heterogeneous effects and show that the implied effect of the boom is larger in states with a younger compulsory schooling age requirement (16 years of age instead of 17 or 18), lower state-level effective tax rate on oil and gas productions, traditional mining, non-metro, and persistent poverty counties.

La richesse naturelle est-elle un gage de prospérité ou se révèle-t-elle être une malédiction? Comment le pétrole a-t-il façonné l'évolution économique des pays producteurs ? Dans le prolongement de ces interrogations, l'objectif de cette thèse est de progresser dans la compréhension des mécanismes qui font que le pétrole est, pour les pays exportateurs, aussi souvent une malédiction qu'une bénédiction. Les travaux empiriques qui constituent notre thèse permettent ainsi de répondre à trois questions distinctes : (i) quelle est la contribution du pétrole en tant que ressource énergétique (ou source d'énergie) au processus de croissance économique ? (ii) quels sont les effets directs et indirects de la dépendance aux revenus pétroliers sur la croissance économique et (iii) la malédiction pétrolière n'est-elle pas une question qui renvoie à la stabilité macroéconomique?Notre analyse met ainsi en évidence les résultats suivants : (i) une richesse pétrolière abondante et la surconsommation de pétrole observée dans une large majorité de pays exportateurs de pétrole contribuent positivement au processus de croissance économique. Ce résultat n'est toutefois valable que sur le court terme. En effet, sur le long terme, la consommation de pétrole s'avère être une conséquence de la croissance économique ; (ii) le pétrole en tant que source de revenus impacte la croissance économique directement et indirectement via ses effets sur le montant et la qualité des dépenses publiques ainsi que sur l'ouverture commerciale. Au regard de ces mécanismes de transmission, nos résultats montrent qu'au-delà d'un certain seuil de dépendance aux revenus pétroliers, la croissance économique est entravée par les effets directs et indirects de la rente pétrolière. Toutefois, ces effets peuvent être contenus, tout d'abord, en réduisant la dépendance aux revenus pétroliers, en améliorant, ensuite, la gouvernance et, enfin, en allant vers davantage de stabilité politique ; (iii) les revenus pétroliers, de part leur extrême instabilité peuvent nuire à la croissance économique en induisant des distorsions macroéconomiques. Cette instabilité se traduit plus précisément par une appréciation du taux de change réel, une hausse des dépenses publiques et de l'inflation. Les pays les plus tributaires de la rente pétrolière sont les plus exposés à cette instabilité macroéconomique. De même, les pays où l'efficacité et la crédibilité du gouvernement sont moindres sont ceux où la croissance économique pâtit le plus de cette instabilité macroéconomique.Le pétrole est ainsi un atout pour les économies des pays exportateurs de pétrole dont il faut maitriser les effets indésirables sur l'économie. Une première solution consisterait alors à réduire le niveau de dépendance de l'économie aux revenus pétroliers pour diminuer le risque d'exposition à la volatilité des prix du pétrole et en réduire le risque de contagion à l'économie. Une autre solution nécessiterait d'améliorer la capacité des gouvernements à mettre en place des politiques économiques efficientes. / Is natural wealth a guarantee of prosperity or is it a curse? How has petroleum shaped growth economic process in oil producing countries? To the extent that these questions have to be raise, the purpose of this thesis is to move towards a better understanding of the mechanisms that make oil becoming a curse as often as a blessing, in oil exporting countries. The empirical studies conducted in this thesis help answer three main questions: (i) What is the contribution of oil as energy (or an energy source) in the process of economic growth? (ii) What are the direct and indirect effects of dependence to oil revenues on economic growth? (iii) Is the oil curse a question of macroeconomic stability?Our contributions thus highlight the following results. (i) Abundant oil wealth and overconsumption observed in the vast majority of oil exporting countries contribute positively to the economic growth process. This result is, however, valid only in the short term. Indeed, in the long term, oil consumption appears to be a consequence of economic growth. (ii) Oil as a source of revenue impacts economic growth directly and indirectly through its effect on the amount and quality of public spending as well as on trade openness. Given these mechanisms, our results show that beyond a certain threshold of dependence on oil revenues, economic growth is constrained by the direct and indirect effects of oil revenues. However, these effects can be contained, first, by reducing dependence on oil revenues; then, by improving government effectiveness; and finally by increasing political stability. (iii) Oil revenues, due to their extreme instability may harm economic growth by inducing macroeconomic distortions. This instability results more precisely by an appreciation of the real exchange rate, a rise in public spending and inflation. The most dependent are countries, the most they are exposed to macroeconomic instability. Similarly, countries with an efficient and credible government are the one which suffer economic growth suffers the less from macroeconomic instability.Oil is, thus, a vantage for oil exporting countries but the adverse effects of such a natural resource on the economy must be mastered. One solution would, then, be to reduce the level of dependence of the economy on oil revenues to reduce the exposure to volatile oil prices and to reduce the risk of contagion to the economy. Another solution would be to improve the ability of governments to implement efficient economic policies.

The resource curse hypothesis suggests that resource-rich countries show lower economic growth rates compared to resource-poor countries. We add to this literature by providing empirical evidence on a new transmission channel of the resource curse, namely, the negative effect of rents on the quality of education. The cross-country analysis for more than 70 countries shows a significantly positive effect of oil rents on the quantity of education measured by government spending on primary and secondary education. Hence, the underspending hypothesis championed by Gylfason (2001) no longer holds with newer data. However, we find a robust and negative effect of oil rents dependency on the current objective and subjective indicators of quality of education, controlling for a set of other drivers of education quality and regional dummies. Despite spending significant shares of GDP on education, oil-rich countries still suffer from an insufficient quality of primary and secondary education, which may hamper their growth potentials. The significant negative effect of oil rents dependency on education quality can be explained by both the demand (e.g., skill acquisition) and supply (e.g., teacher quality) side channels.