The initial public offering (IPO) underpricing phenomenon has frequently been noticed and generally is accepted as a puzzle in financial economics. Some of the new theories, such as behavioural finance, take the underpricing puzzle as one important form of evidence. However, some aspects of IPO underpricing have not yet been fully documented and discussed in the existing literature. This thesis tries to contribute in the following three specific areas. First, we focus on the time series properties of the level of underpricing of IPO shares and document the IPO market in the Hong Kong market from 1999 to 2005. In the data sample, strong autocorrelation within the level of underpricing has been discovered. Evidence suggests the initial selling volume plays an important role in the relationship. The links between underpricing and clustering of IPOs within different industries are weak, suggesting the reasons for underpricing are related to the market liquidity rather than to the industry-specific risk characteristics. Second, we investigate the underwriting networks to explore the relationship between underwriting business and IPO related puzzles. We find that in repeated IPOs, underwriters build up reputation and accumulate knowledge of their underwriting services. One of the great advantages of the top ranked underwriters is their relationship networks with other underwriters and institutional investors. We perform a careful examination of the underwriter syndicate and investigate the relationship of the structure of the syndicate in respect of IPO performance. Moreover, the pattern of distribution in the size of syndicates is identified and is found to be significantly related to the IPO performance. The research shows that the perspective from the underwriter syndicate is not only interesting, also necessary to understand IPOs. Third, we analyse the coordination problem in the IPO. In the research, we consider the auction method as a one-stage selling and the bookbuilding method as a two-stage selling method. The model suggests that the relationship between the underpricing level and the quality of IPO shares is non-monotone. This implication is consistent with empirical observations. In addition, regarding the issuers' proceeds in the IPOs, the auction method is better than the bookbuilding method in both noisy and noisy vanishing equilibria. The bookbuilding method may be helpful in other ways, such as maintaining liquidity or price support in secondary market. By studying liquidity, business networks and the coordination problem, the thesis does not only complement the existing research by providing unique explanations for the IPO underpricing and other related puzzles, but also opens some interesting venues for future research.

Recommender systems automate the process of recommending products and services to customers based on various types of data including customer demographics, product features, and, most importantly, previous interactions between customers and products (e.g., purchasing, rating, and catalog browsing). Despite significant research progress and growing acceptance in real-world applications, two major challenges remain to be addressed to implement effective e-commerce recommendation applications. The first challenge is concerned with making recommendations based on sparse transaction data. The second challenge is the lack of a unified framework to integrate multiple types of input data and recommendation approaches.This dissertation investigates graph-based algorithms to address these two problems. The proposed approach is centered on consumer-product graphs that represent sales transactions as links connecting consumer and product nodes. In order to address the sparsity problem, I investigate the network spreading activation algorithms and a newly proposed link analysis algorithm motivated by ideas from Web graph analysis techniques. Experimental results with several e-commerce datasets indicated that both classes of algorithms outperform a wide range of existing collaborative filtering algorithms, especially under sparse data. Two graph-based models that enhance the simple consumer-product graph were proposed to provide unified recommendation frameworks. The first model, a two-layer graph model, enhances the consumer-product graph by incorporating the consumer/product attribute information as consumer and product similarity links. The second model is based on probabilistic relational models (PRMs) developed in the relational learning literature. It is demonstrated with e-commerce datasets that the proposed frameworks not only conceptually unify many of the existing recommendation approaches but also allow the exploitation of a wider range of data patterns in an integrated manner, leading to improved recommendation performance.In addition to the recommendation algorithm design research, this dissertation also employs the random graph theory to study the topological characteristics of consumer-product graphs and the fundamental mechanisms that generate the sales transaction data. This research represents the early step towards a meta-level analysis framework for validating the fundamental assumptions made by different recommendation algorithms regarding the consumer-product interaction generation process and thus supporting systematic recommendation model/algorithm selection and evaluation.

Barabási-Albert random graph models are a class of evolving random graphs that are frequently used to model social networks with scale-free degree distributions. It has been shown that Barabási-Albert random graph models have asymptotic scale-free degree distributions as the size of the graph tends to infinity. Real world networks, however, have finite size so it is important to know how close the degree distribution of a Barabási-Albert random graph of a given size is to its asymptotic distribution. Stein’s method is chosen as one main method for obtaining explicit bounds for the distance between distributions. We derive a new version of Stein’s method for a class of scale-free distributions and apply the method to a Barabási-Albert random graph. We compare the evolution of a sequence of Barabási-Albert random graphs with continuous time stochastic processes motivated by Yule’s model for evolution. Through a coupling of the models we bound the total variation distance between their degree distributions. Using these bounds, we extend degree distribution bounds that we find for specific models within the scheme to find bounds for every member of the scheme. We apply the Azuma-Hoeffding inequality and Chernoff bounds to find bounds between the degree sequences of the random graph models and the given scale-free distribution. These bounds prove that the degree sequences converge completely (and therefore also converge almost surely) to our scale-free distribution. We discuss the relationship between the random graph processes and the Chinese restaurant process. Aided by the construction of an inhomogeneous Markov chain, we apply our results for the degree distribution in a Barabási-Albert random graph to a particular statistic of the Chinese restaurant process. Finally, we explore how our methods can be adapted and extended to other evolving random graph processes. We study a Bernoulli evolving random graph process, for which we bound the distance between its degree distribution and a geometric distribution and we bound the distance between the number of triangles in the graph and a normal distribution.

In 2011, Oregon was one of many states in the U.S. consolidating their education governance around an early learning, K-12, and postsecondary hub. This study uses legislator-voting data to investigate the relationship between this consolidated model and endogenous policy formulation processes. This study employs a separable temporal exponential random graph model (STERGM) to investigate how an education governance shift toward consolidated authority relates to bipartisan outcomes for education-related bills over time. Oregon legislator voting networks were analyzed for cohesion, centrality, and community detection measures, as well as by legislator attributes (e.g. gender, party, and title) to test the association they had on the likelihood of forming ties with other legislators. Finally, to study the relationship of bipartisanship with legislators’ likelihood to vote commonly, I added the legislators’ political party attributes within dyads to analyze the association that having different political parties had on legislators’ common votes. The results highlight evidence of legislator networks that were very dense at each time point included in the study, with a high likelihood of forming ties. However, when Oregon shifted to centralized education governance model their legislator networks became more distributed and cohesive when compared to other years included in the longitudinal study. It is possible that such a shift prompted collaboration among legislators resulting in mutuality that increased the likelihood for underrepresented groups of legislators (e.g. females and republicans) to vote commonly with their colleagues. Aligned with previous research, this study found that centralized governing bodies reinforced by political legislation provided collaborative initiatives for the legislative community. Attending to bipartisan voting patterns dynamically through a governance shift is a valuable investigation that will provide nuanced inferences about education governance and policymaking for states making similar consolidated governance shifts in the future.

This research seeks an answer to the following question: what is the relationship between the structure of researchers' communication network and the structure of their collaborative output networks (e.g. co-authored publications, joint grant proposals, and joint patent applications), and the impact of these structures on their citation performance and the volume of collaborative research outputs? Three complementary studies are performed to answer this main question as discussed below. 1. Study I: A frequently used output to measure scientific (or research) collaboration is co-authorship in scholarly publications. Less frequently used are joint grant proposals and patents. Many scholars believe that co-authorship as the sole measure of research collaboration is insufficient because collaboration between researchers might not result in co-authorship. Collaborations involve informal communication (i.e., conversational exchange) between researchers. Using self-reports from 100 tenured/tenure-track faculty in the College of Engineering at the University of South Florida, researchers' networks are constructed from their communication relations and collaborations in three areas: joint publications, joint grant proposals, and joint patents. The data collection: 1) provides a rich data set of both researchers' in-progress and completed collaborative outputs, 2) yields a rating from the researchers on the importance of a tie to them 3) obtains multiple types of ties between researchers allowing for the comparison of their multiple networks. Exponential Random Graph Model (ERGM) results show that the more communication researchers have the more likely they produce collaborative outputs. Furthermore, the impact of four demographic attributes: gender, race, department affiliation, and spatial proximity on collaborative output relations is tested. The results indicate that grant proposals are submitted with mixed gender teams in the college of engineering. Besides, the same race researchers are more likely to publish together. The demographics do not have an additional leverage on joint patents. 2. Study II: Previous research shows that researchers' social network metrics obtained from a collaborative output network (e.g., joint publications or co-authorship network) impact their performance determined by g-index. This study uses a richer dataset to show that a scholar's performance should be considered with respect to position in multiple networks. Previous research using only the network of researchers' joint publications shows that a researcher's distinct connections to other researchers (i.e., degree centrality), a researcher's number of repeated collaborative outputs (i.e., average tie strength), and a researchers' redundant connections to a group of researchers who are themselves well-connected (i.e., efficiency coefficient) has a positive impact on the researchers' performance, while a researcher's tendency to connect with other researchers who are themselves well-connected (i.e., eigenvector centrality) had a negative impact on the researchers' performance. The findings of this study are similar except that eigenvector centrality has a positive impact on the performance of scholars. Moreover, the results demonstrate that a researcher's tendency towards dense local neighborhoods (as measured by the local clustering coefficient) and the researchers' demographic attributes such as gender should also be considered when investigating the impact of the social network metrics on the performance of researchers. 3. Study III: This study investigates to what extent researchers' interactions in the early stage of their collaborative network activities impact the number of collaborative outputs produced (e.g., joint publications, joint grant proposals, and joint patents). Path models using the Partial Least Squares (PLS) method are run to test the extent to which researchers' individual innovativeness, as determined by the specific indicators obtained from their interactions in the early stage of their collaborative network activities, impacts the number of collaborative outputs they produced taking into account the tie strength of a researcher to other conversational partners (TS). Within a college of engineering, it is found that researchers' individual innovativeness positively impacts the volume of their collaborative outputs. It is observed that TS positively impacts researchers' individual innovativeness, whereas TS negatively impacts researchers' volume of collaborative outputs. Furthermore, TS negatively impacts the relationship between researchers' individual innovativeness and the volume of their collaborative outputs, which is consistent with `Strength of Weak Ties' Theory. The results of this study contribute to the literature regarding the transformation of tacit knowledge into explicit knowledge in a university context.

Wireless Networks have been the topic of fundamental research in recent years with the aim of achieving reliable and efficient communications. However, due to their complexity, there are still many aspects of such configurations that remain as open problems. The focus of this thesis is to investigate some throughput limits of wireless networks. The network under consideration consists of $n$ source-destination pairs (links) operating in a single-hop fashion. In Chapters 2 and 3, it is assumed that each link can be active and transmit with a constant power P or remain silent. Also, fading is assumed to be the dominant factor affecting the strength of the channels between transmitter and receiver terminals. The objective is to choose a set of active links such that the throughput is maximized, where the rate of active links are either unconstrained or constrained. For the unconstrained throughput maximization, by deriving an upper bound and a lower bound, it is shown that in the case of Rayleigh fading: (i) the maximum throughput scales like $\log n$, (ii) the maximum throughput is achievable in a distributed fashion. The upper bound is obtained using probabilistic methods, where the key point is to upper bound the throughput of any random set of active links by a chi-squared random variable. To obtain the lower bound, a threshold-based link activation strategy (TBLAS) is proposed and analyzed. The achieved throughput of TBLAS is by a factor of four larger than what was obtained in previous works with centralized methods and with multihop communications. When the active links are constrained to transmit with a constant rate $\lambda$, an upper bound is derived that shows the number of active links scales at most like $\frac{1}{\lambda} \log n$. It is proved that TBLAS \emph{asymptotically almost surely(a.a.s.)} yields a feasible solution for the constrained throughput maximization problem. This solution, which is suboptimal in general, performs close to the upper bound for small values of $\lambda$. To improve the suboptimal solution, a double-threshold-based link activation strategy (DTBLAS) is proposed and analyzed based on some results from random graph theory. It is demonstrated that DTBLAS performs very close to the optimum. Specifically, DTBLAS is a.a.s. optimum when $\lambda$ approaches $\infty$ or $0$. The optimality results are obtained in an interference-limited regime. However, it is shown that, by proper selection of the algorithm parameters, DTBLAS also allows the network to operate in a noise-limited regime in which the transmission rates can be adjusted by the transmission powers. The price for this flexibility is a decrease in the throughput scaling law by a factor of $\log \log n$. In Chapter 4, the problem of throughput maximization by means of power allocation is considered. It is demonstrated that under individual power constraints, in the optimum solution, the power of at least one link should take its maximum value. Then, for the special case of $n=2$ links, it is shown that the optimum power allocation strategy for throughput maximization is such that either both links use their maximum power or one of them uses its maximum power and the other keeps silent.

Wireless Networks have been the topic of fundamental research in recent years with the aim of achieving reliable and efficient communications. However, due to their complexity, there are still many aspects of such configurations that remain as open problems. The focus of this thesis is to investigate some throughput limits of wireless networks. The network under consideration consists of $n$ source-destination pairs (links) operating in a single-hop fashion. In Chapters 2 and 3, it is assumed that each link can be active and transmit with a constant power P or remain silent. Also, fading is assumed to be the dominant factor affecting the strength of the channels between transmitter and receiver terminals. The objective is to choose a set of active links such that the throughput is maximized, where the rate of active links are either unconstrained or constrained. For the unconstrained throughput maximization, by deriving an upper bound and a lower bound, it is shown that in the case of Rayleigh fading: (i) the maximum throughput scales like $\log n$, (ii) the maximum throughput is achievable in a distributed fashion. The upper bound is obtained using probabilistic methods, where the key point is to upper bound the throughput of any random set of active links by a chi-squared random variable. To obtain the lower bound, a threshold-based link activation strategy (TBLAS) is proposed and analyzed. The achieved throughput of TBLAS is by a factor of four larger than what was obtained in previous works with centralized methods and with multihop communications. When the active links are constrained to transmit with a constant rate $\lambda$, an upper bound is derived that shows the number of active links scales at most like $\frac{1}{\lambda} \log n$. It is proved that TBLAS \emph{asymptotically almost surely(a.a.s.)} yields a feasible solution for the constrained throughput maximization problem. This solution, which is suboptimal in general, performs close to the upper bound for small values of $\lambda$. To improve the suboptimal solution, a double-threshold-based link activation strategy (DTBLAS) is proposed and analyzed based on some results from random graph theory. It is demonstrated that DTBLAS performs very close to the optimum. Specifically, DTBLAS is a.a.s. optimum when $\lambda$ approaches $\infty$ or $0$. The optimality results are obtained in an interference-limited regime. However, it is shown that, by proper selection of the algorithm parameters, DTBLAS also allows the network to operate in a noise-limited regime in which the transmission rates can be adjusted by the transmission powers. The price for this flexibility is a decrease in the throughput scaling law by a factor of $\log \log n$. In Chapter 4, the problem of throughput maximization by means of power allocation is considered. It is demonstrated that under individual power constraints, in the optimum solution, the power of at least one link should take its maximum value. Then, for the special case of $n=2$ links, it is shown that the optimum power allocation strategy for throughput maximization is such that either both links use their maximum power or one of them uses its maximum power and the other keeps silent.

Knowledge of the terms sex and gender has important political, health and equity considerations. This thesis investigates the macrostructural assertions of Connell’s social theory of gender which is fundamentally concerned with demonstrating the relational and hierarchical nature of gender. A major criticism of the theory has been its lack of account of the individual and the ways in which gender is performed in local settings. Therefore, investigation primarily concerns whether Connell’s macrostructural theory is explicable in local social contexts. A theoretical critique and reframing of the theory lead to articulating the interdependency between structural, cultural and individual factors. By explicitly stating Connell’s implicit assertions, what becomes evident is that people’s gendered relations are interrelated with beliefs which are held personally and shared culturally. Specifically, a major theoretical impasse is overcome when recognising that the “ideology of supremacy” of a dominative masculinity is necessarily interdependent with the structural relations of power. / From here I have suggested that there are particular patterns of these structures and beliefs which can be seen in macrostructural terms, but also in local settings. These hypotheses are reframed into social network terms for an empirical investigation of Connell’s theory in local contexts. To determine whether the predicted hypotheses for Connell’s theory occur at greater than chance levels, a particular type of statistical model for social networks, called exponential random graph (p*) models, is employed. Importantly, such models utilize a methodological approach which specifically acknowledges the interdependency of structural, individual and cultural factors, thus aligning Connell’s theory with the method of investigation. / Primarily, Connell’s theory is concerned with differing configurations of masculinity, and for this reason my focus is predominantly on males and their relations with one another. To this end, two general local settings were chosen to explore these issues – secondary schools and all-male elite-level (AFL) sporting teams. Social network models were used to examine the relations between masculinities in six schools and four AFL clubs. Importantly, Connell has suggested that local contexts are likely to differ from one another in the degree to which they support gendered relations of power. Results for schools and clubs vary considerably from one another in the ways in which they provide local level support for Connell’s theory. Significantly though, there are some contexts which do show support for Connell’s theory. That such evidence can be found to endorse specifically defined local-level predictions for Connell’s theory, even when controlling for complex micro-level social structures, and also for other individual level effects, and still produce statistically significant effects supporting these predictions suggests that support is not trivial. There is strong evidence that attitudes towards masculinity can be an important organising principle in the emergence of hierarchy, not universally, but in some contexts. / It can be concluded that gender relations tend to operate in ways predicted by Connell’s theory, though local context is particularly important. The specific findings from local social contexts do have wider implications for Connell’s theory, including how hierarchy in gender can be tied to other structures of power, where femininities fit into the theory, a more precise account of hegemony and an exploration of the impact culture has in local settings.

abstract: The cyber-physical systems (CPS) are emerging as the underpinning technology for major industries in the 21-th century. This dissertation is focused on two fundamental issues in cyber-physical systems: network interdependence and information dynamics. It consists of the following two main thrusts. The first thrust is targeted at understanding the impact of network interdependence. It is shown that a cyber-physical system built upon multiple interdependent networks are more vulnerable to attacks since node failures in one network may result in failures in the other network, causing a cascade of failures that would potentially lead to the collapse of the entire infrastructure. There is thus a need to develop a new network science for modeling and quantifying cascading failures in multiple interdependent networks, and to develop network management algorithms that improve network robustness and ensure overall network reliability against cascading failures. To enhance the system robustness, a "regular" allocation strategy is proposed that yields better resistance against cascading failures compared to all possible existing strategies. Furthermore, in view of the load redistribution feature in many physical infrastructure networks, e.g., power grids, a CPS model is developed where the threshold model and the giant connected component model are used to capture the node failures in the physical infrastructure network and the cyber network, respectively. The second thrust is centered around the information dynamics in the CPS. One speculation is that the interconnections over multiple networks can facilitate information diffusion since information propagation in one network can trigger further spread in the other network. With this insight, a theoretical framework is developed to analyze information epidemic across multiple interconnecting networks. It is shown that the conjoining among networks can dramatically speed up message diffusion. Along a different avenue, many cyber-physical systems rely on wireless networks which offer platforms for information exchanges. To optimize the QoS of wireless networks, there is a need to develop a high-throughput and low-complexity scheduling algorithm to control link dynamics. To that end, distributed link scheduling algorithms are explored for multi-hop MIMO networks and two CSMA algorithms under the continuous-time model and the discrete-time model are devised, respectively. / Dissertation/Thesis / Ph.D. Electrical Engineering 2012

How are arms export choices made within a state? In this dissertation I use a foreign policy analysis framework to examine this question. I focus on examining each of the three primary levels of analysis in international relations as it relates to the main question. I begin with a typical international relations level and examine the characteristics of the two states that dominate the world arms trade: The United States and Russia. I then examine the full network of relations among all states in the international system that are involved in the sale or purchase of arms. To do this I use an Exponential Random Graph Model (ERGM) to examine these relations, which I derived from data on arms sales from the Stockholm International Peace Research Institute (SIPRI). I examine the arms sales in each decade from 1950 through 2010. In order to answer the question of how arms decisions are made within the state, I focus my inquiry on the United States and Russia. It is these states that have the practical capability to use arms transfers as a foreign policy tool. I examine the foreign policy making mechanisms in each of these states to determine how arms transfers can be used as a foreign policy tool. I examine and the bureaucratic institutions within each state and come up with a state ordering preference for how arms decisions are evaluated in each state. Finally, I use case studies to examine arms relations between the both the U.S. and Russia and three other states in each case. The other states were selected based on the pattern of sales between the two countries. I examine these sales to determine the impact of bureaucratic maneuvering and interest politics on the decision-making process within Russia and the United States. I find in my network analysis that the traditional measures of state power - military spending, regime type, and military alliances - do not account for the overall structure of the arms sale network. The most important factors in the formation of the arms sale network in each of the six decades that I study are specific configurations of triadic relations that suggest a continued hierarchy in the arms sale network. I find in my case study chapters that a simple model of state interest as a decision-making rule accounts for the decisions made by the different bureaucratic actors in the U.S. Russian arms sales are driven by a state imperative to increase Russia's market share, and there is high-level involvement in making different arms deals with other countries.