This dissertation project broadly addresses the question of how state behavior is conditioned by the structural configuration of the network in which they are embedded. It attempts to reconcile some of the discrepancies between the systemic and dyadic approaches to international relations, by arguing that the international system is a multidimensional network that results as an emergent property of the dyadic ties that exist between states.This dissertation consists of three stand alone analyses, connected by their focus on systemic configuration and the impact of various elements of international structures on the behavior of states. In contrast to most studies of international relations, dimensions of the international order are observed, rather than assumed. The first chapter focuses specifically on observing and describing the structure and tendencies of the behavioral dimension of the international system. It assesses patterns of consistency in international relations, searching for both simple, dyadic consistency as well as more complex, triadic consistency. The second chapter relies on these positive and negative relations to create a model of policy reinforcement, with a focus on the onset of militarized conflict. Structural balance theory is used to identify shared, external relations that either reinforce or dampen the impact of dyadic hostility on militarized conflict. The final empirical chapter shifts to a more localized focus, investigating the impact of alliance portfolio size on the likelihood of alliance obligation fulfillment. Here, the size and capabilities of a disputant's local alliance portfolio do not directly modify the behavior of the disputant, but instead that of the disputant's other allies.Together, these chapters demonstrate the importance of accounting for systemic factors in explaining and analyzing dyadic behavior. The characteristics of local networks, such as alliance portfolios, have significant implications for state security. The configuration of foreign policy relations provides feedback to states, influencing their willingness to take aggressive actions. Bipolarity and multipolarity can be empirically observed through the clustering of states, rather than by merely counting the number of major powers. However, perhaps most significant is the contribution these analyses make to a small but growing literature attempting to move beyond the dyad.

Collaborative groups are important both in the learning environment of engineering education and, in the real world, the business of engineering design. Selecting appropriate individuals to form an effective group and monitoring a group’s progress are important aspects of successful task performance. This exploratory study looked at using the concepts of cognitive social structures, structural balance, and centrality from social network analysis as well as the measures of emotional intelligence. The concepts were used to analyze potential team members to examine if an individual's ability to perceive emotion in others and the self and to use, understand, and manage those emotions are a factor in a group’s performance. The students from a capstone design course in computer engineering were used as volunteer subjects. They were formed into groups and assigned a design exercise to determine whether and which of the above mentioned tools would be effective in both selecting teams and predicting the quality of the resultant design. The results were inconclusive with the exception of an individual's ability to accurately perceive emotions. The instruments that were successful were the Self-Monitoring scale and the accuracy scores derived from cognitive social structures and Level IV of network levels of analysis.

This thesis targets at providing an algebraic method to indicate network behaviors. Furthermore, for a signed-average consensus problem of the system behaviors, event-triggering signed-average algorithms are designed to reduce the communication overheads. In Chapter 1, the background is introduced, and the problem is formulated. In Chapter 2, notations and basics of graph theory are presented. It is known that the terminal value of the system state is determined by the initial state, left eigenvector and right eigenvector associated with zero eigenvalue of the Laplacian matrix. Since there is no mathematical expression of right eigenvector, in Chapter 3, mathematical expression of right eigenvector is given. In Chapter 4, algebraic trait for structurally balanced property of a node is proposed. In Chapter 5, a method for characterization of collective behaviors under directed signed networks is developed. In Chapter 6, dynamic event-triggering signed-average algorithms are proposed and proved for the purpose of relieving the communication burden between agents. Chapter 7 summarizes the thesis and gives future directions.

Submitted by Celso Magalhaes (celsomagalhaes@ufrrj.br) on 2018-08-28T12:21:13Z No. of bitstreams: 1 2017 - Victor dos Santos Ponciano.pdf: 551672 bytes, checksum: c01424f760ef08254a63d236b62f72c5 (MD5) / Made available in DSpace on 2018-08-28T12:21:15Z (GMT). No. of bitstreams: 1 2017 - Victor dos Santos Ponciano.pdf: 551672 bytes, checksum: c01424f760ef08254a63d236b62f72c5 (MD5) Previous issue date: 2017-02-21 / Conselho Nacional de Desenvolvimento Cient?fico e Tecnol?gico - CNPq / In 1946, Heider developed Signal Graph Theory with the purpose of describing the emotional relationships between people pertaining to the same social group. In this work, we study graph partitioning problems associated with structural balance. These problems are known in the computer science literature as partition correlation problems: correlation clustering (CC) and a relaxed version (RCC). The solution of CC and RCC problems has been previously used in the literature as a tool for the evaluation of structural balance in a social network. The aim of this work is to apply the solution of these problems in the detection of corruption in public contracts. We describe integer linear programming formulations from the literature for these problems. We also discuss a probabilistic model for Structural balance and the solution of these problems applied to the detection of cases of corruption existing in public bids. / Em 1946, Heider desenvolveu a Teoria de Grafos de Sinais a fim de descrever as rela??es emocionais entre as pessoas pertencentes ao mesmo grupo social. Neste trabalho, estudamos problemas de particionamento de grafo associados com equil?brio estrutural, que na literatura de Ci?ncia da Computa??o s?o conhecidos como problemas de correla??o de parti??es ou, em ingl?s, correlation clustering (CC), al?m de uma vers?o relaxada (RCC). As solu??es dos problemas CC e RCC foram anteriormente utilizadas na literatura como ferramentas para a avalia??o de equil?brio estrutural numa rede social. O objetivo deste trabalho ? aplicar as solu??es destes problemas na detec??o de corrup??o em contratos p?blicos. Para esse fim, s?o utilizadas formula??es de programa??o linear inteira existentes na literatura para estes problemas. Al?m disso, ? discutido tamb?m um modelo probabil?stico para o Equil?brio Estrutural e as solu??es destes problemas s?o aplicadas ? detec??o de casos de corrup??es existentes em licita??es p?blicas.

Submitted by Silvana Teresinha Dornelles Studzinski (sstudzinski) on 2016-10-05T12:57:37Z No. of bitstreams: 1 Bruno Pedron_.pdf: 427411 bytes, checksum: 3ec46db7e95778a77ba0a7da26faffca (MD5) / Made available in DSpace on 2016-10-05T12:57:37Z (GMT). No. of bitstreams: 1 Bruno Pedron_.pdf: 427411 bytes, checksum: 3ec46db7e95778a77ba0a7da26faffca (MD5) Previous issue date: 2016-07-21 / itt Performance - Instituto Tecnológico em Desempenho da Construção Civil / O objetivo deste trabalho foi compreender a relação entre o ciclo de negócios e o resultado primário do Rio Grande do Sul. Para tanto, foram coletados dados mensais da Ferramenta de finanças públicas em tempo real da Secretaria da Fazenda do estado para o período de 1999-2015, os quais conduziram ao cálculo do resultado primário trimestral sem operações intraorçamentárias. Estes dados foram ajustados pela inflação pelo IPCA e pela sazonalidade utilizando o Census X-13 multiplicativo, e então foram utilizados para estimar o balanço orçamentário estrutural pela abordagem do Fundo Monetário Internacional. As repartições de receitas da União e as transferências diversas foram removidas da regressão e a elasticidade das receitas em relação ao PIB foi estimada através de um modelo autorregressivo, resultando em 1,0908, enquanto o PIB potencial foi obtido pelo uso do filtro Christiano-Fitzgerald. Nenhum ajuste foi aplicado às despesas ou às transferências da União. Os resultados mostraram que o Rio Grande do Sul está mais predisposto a operar acima do PIB potencial e promovendo contrações fiscais, apesar de mostrar mais esforços nas fases de expansão. Assim, foram encontradas políticas pró-cíclias em 75% dos anos da amostra. Quando os ciclos políticos foram analisados, apenas o ciclo da governadora Yeda Crusius (2007-2010) apresentou-se de forma anti-cíclica. De forma geral, o estado está alinhado à ciclicidade do governo federal, mesmo considerando que, metade dos períodos que foram comparados com Andreis (2014) apresentaram políticas opostas às estimadas. Finalmente sugestões foram dadas para o estado na forma de contrações fiscais, inspirado em Santos (2014). / The aim of this study is to understand the relation between the business cycle and the primary result of Rio Grande do Sul. For that reason, monthly data were collected from the Real-time public finances tool of the state finances office for the period of 1999-2015, which allowed for the calculation of the quarterly primary result without intra-budget operations. These data were adjusted for inflation by IPCA and seasonality by multiplicative Census X-13 and then used to run the International Monetary Fund (IMF) approach of the structural budget balance. The transfers from the federal government were removed from the regression and the elasticity of revenue to the output was estimated through an autoregressive model and resulted in 1.0908, while the potential GDP was obtained through the Christiano-Fitzgerald filter. No adjustments were applied to the expenditures or the transfers. The results pointed out that Rio Grande do Sul is more likely to be operating above trend GDP and taking actions of fiscal contraction, although working harder during phases of expansion. Thus, pro-cyclical policies were conducted on 75% of the years of the sample. When the political cycles were analyzed, it turned out only the cycle of governor Yeda Crusis (2007-2010) had a counter-cyclical bias. Overall, the state is aligned with the cyclicality of the federal government, even though half of the periods that were compared to the findings of Andreis (2014) displayed opposite fiscal stances. In the end, some suggestions were made for fiscal contractions, inspired by Santos (2014).

The present thesis is concerned with the existence of political cycles in European Union member states between 1990 and 2013. These cycles are of two types: political business cycles and political budget cycles. The analysis was performed by means of the fixed effects method (using first differences) along with a visual analysis of data. The results obtained from three types of data sets show that two years before elections, inflation grows at 0.47%, unemployment rate at 0.5%, and structural balance at 0.8%. One of the data sets reveals that structural balance is reduced by 0.39% in election years. The visual analysis clearly demonstrates that there exist political business cycles in Belgium, Bulgaria, Czech Republic, Denmark, Germany, Poland and Sweden. Political budget cycles have been proven to exist in Belgium, France, Cyprus, Malta, Germany, Slovakia and the United Kingdom.

After many decades, macroeconomic effects of fiscal policy have returned to the centre of the economic policy debate. Both automatic fiscal stabilizers and discretionary fiscal stimuli have been used to support aggregate demand during the recent global economic crisis with a subsequent need for large-scale fiscal consolidations. In this context, a proper assessment of the size of automatic fiscal stabilizers and fiscal multipliers represents a key input for fiscal policymaking. This dissertation provides a quantitative analysis of the interactions between fiscal policy and real economy in the Czech Republic. The impact of real economy developments on public finances is assessed based on the methods of the OECD, the European Commission and the ESCB for the identification of general government structural balances, i.e. balances adjusted for effects of the economic cycle and net of one-off and other temporary transactions. I find that the underlying fiscal position, as approximated by the government structural balance, was mostly below the level stabilising the debt-to-GDP ratio since mid-1990s. An indistinct improvement in the structural balance can be identified in the period 2004--2007, which was subsequently reversed by the adverse structural impact of the world economic crisis. At the same time, dynamics of unadjusted fiscal balance was largely determined by one-off transactions in the past. The effects of fiscal policy on real economy are analysed using the structural VAR approach. I find that an increase in government spending has a temporary positive effect on output that peaks after one to two years with a multiplier of around 0.6. Tax multiplier appears to be small and, in contrast to standard Keynesian assumptions, positive. Government spending is supportive to private consumption, contradicting the hypothesis of Ricardian equivalence, but it crowds out private investment in the short run. The results should be interpreted with caution, as the analysis is complicated by rapidly changing economic environment in the period of the economic transition, relatively short available time series and a large number of one-off fiscal transactions.