Network theory

About: Network theory is a research topic. Over the lifetime, 2257 publications have been published within this topic receiving 109864 citations.

Network properties and evolutionof the Hungarian RTGSover the past decade

Abstract: Since the 2008 economic crisis, network research has become increasingly prominent in the world of finance. The complex interrelations and financial interdependencies formed among financial market participants have proved to be critical in times of crisis. In this paper, we explore the network properties of the Hungarian RTGS (VIBER) and also seek an answer to the question of whether the network properties of the system have changed over the long term across the time windows considered, and if so, to what extent. Furthermore, we identify systemically important participants using a variety of network theory tools. We also explore methodologies which – by providing new perspectives for monitoring the evolution of systemically important participants – may contribute to improving the effectiveness of oversight in Hungary. To identify systemically important participants, we apply four methodologies, namely: the LSI index, the model capturing the relation of eigenvector and betweenness, diffusion centrality, and the model exploring the effect of combining multiple nodes. In the Hungarian RTGS, two distinct groups emerge: the first is comprised of participants that play a key role in the transmission of liquidity (“core”), and the other is the cluster of periphery participants. As the composition of the core has remained virtually unchanged, it can be considered stable. While the risk of contagion arising from an operational disruption increased at both the individual and aggregated level during the period under review, it is also apparent that no such link exists in the graph the removal of which would ultimately cut the communication between the banks originally connected by it. The results of each indicator showed that there were no significant changes regarding the network properties across the three time windows, confirming the robustness of the network properties of the Hungarian RTGS and its stability over time.

Effect Analysis of an Additional Edge on Centrality and Ranking of Graph Using Computational Experiments

Abstract: The centrality is calculated to describe the importance of a node in a graph and ranking is given according to the centrality for each node. There are many centrality measures and we use degree centrality, closeness centrality, eigenvector centrality, and betweenness centrality. In this paper, we analyze the effect of an additional edge of a graph on centrality and ranking through experimental computations. It is found that the effect of an additional edge on centrality and ranking of the nodes in the graph is different according to the graph structure using PCA. The results can be used for define the graph characteristics.

Topology and strategy for supply network robustness

Abstract: Qualitative prescriptions for the robustness of supply networks provides individual firms with useful heuristics, but the focus on specific firms and their immediate supplier relationships means that systemic effects that arise in the larger, complex network cannot be taken into account. In this paper we contribute concepts from network theory to relate topological properties of supply networks to the robustness of the system as a whole. To do so, we model a network that has been validated by two companies with global supply chains, and use computer simulations to explore the extent to which firm failures disrupt the network. Our results show that outsourcing, adaptive capacity, and redundant capability have positive effects on robustness whereas alignment of manufacturing strategies does not. We also find that when supply chains are formed through preferential attachment they become more vulnerable to the failure of firms with many links, but they also become more robust financially, as forecasting becomes more accurate. Further research needs to focus on cross-validation with empirical data including observations of temporary production re-allocation, so that we can move beyond models in which we are restricted to a binary choice between failure and optimality.

Structural Analysis of the Overall Hypoxia Response Network

Abstract: The mechanism of hypoxia is an important research topic of biomedicine. The current studies usually only focus on particular aspects of hypoxia-specific issues based on methods and technologies of traditional biomedicine, and seldom investigate from a systematic perspective. In this paper, the mechanism of hypoxia is treated from a network point of view. The structural composition of the hypoxia response network is analyzed from the perspective of complex networks. The overall network of the HIF-1 signaling pathway is established, and its properties, such as the topology, centrality, and cohesive subgroups are analyzed based on network theory and analytical methods. Based on these network properties, the biomedical implications of the corresponding network indicators are explained, the critical nodes and the core components of the network of HIF-1 signaling pathway are obtained, and the effectiveness of the network analytical method is validated.