Democritus University of Thrace (DUTH)
A graph theory approach of the study of macroeconomics networks
Abstract
dc:descriptionMany modern economic systems are characterized by an increased degree of complexity. The interacting agents of these systems develop individual emergent and non-linear behaviors that cannot be fully described by econometric techniques. In recent years, due to the fast increase of computational power and the evolution of algorithms, the science of Network Analysis has been integrated in the analysis of such complicated economic systems to complement the use of econometrics. A commonly used technique in the context of Network Analysis is the Minimum Spanning Tree (MST). The MST produces a sub-graph of the initial network in which all the nodes are connected so that no loops exist. However, the MST bears some inherent drawbacks that stem directly from its algorithmic identification process and may render it inappropriate for the study of economics networks. This dissertation aims to pin-point the disadvantages of the MST when used in economics networks and to highlight the advantages of a new optimization technique, called the Threshold-Minimum Dominating Set (T-MDS), as a more appropriate solution. Furthermore, the Threshold Weighted - Minimum Dominating Set (TW-MDS) is introduced, which sustains all the advantages of the T-MDS and, depending on the data set at hand, it may be more suitable for inter-temporal analyses that are performed across time. The superiority of the T-MDS and TW-MDS over the classic MST is initially displayed in this dissertation with appropriate theoretical examples. We then continue by delivering a diverse set of macroeconomic applications: business cycle synchronization, income inequality evolution and core inflation measurement. By doing this we show the suitability of the proposed methodologies in macroeconomic analysis. Thus, this dissertation has twofold contributions to the analysis of Complex Economics Networks: on the theoretical side it advances the relative literature by providing a more appropriate tool than the one used so far, while on the empirical side it delivers new insights from the diverse economic applications of the T-MDS.
Degree
thesis:*- Grantor dc:publisher
- Democritus University of Thrace (DUTH)
- Year dc:date
- 2017
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Sarantitis, Georgios-Antonios
- Σαραντίτης, Γεώργιος-Αντώνιος
Subjects
dc:subject × 8Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/41291