University of the Aegean
Analysing investment patterns in financial markets using social network analysis
Abstract
dc:descriptionWe investigate shareholder activism networks from 1999 to 2018 and find the presence of the small-world phenomenon, which fosters the participation of shareholders and public firms in shareholder activism. The small-world phenomenon is also associated with the presence of brokers, which are nodes that bridge shareholder activists with other shareholder activists and targets. Moreover, the shareholder activism network is shaped by star nodes, that is, nodes that have many adjacent nodes, most of which are not connected to each other. Although star nodes affect the presence of the small-world phenomenon, there is weak evidence in support of the association of star nodes with the participation of shareholders and public firms in shareholder activism or with the probability of brokers’ presence. Furthermore, we find that cash-on-hand and firm size seem to have the most significant impact on the probability of brokers’ presence in the shareholder activism network. Moreover, this thesis seeks to detect herding behaviour in public firms targeted by multiple shareholders through shareholder activism. We use beneficial ownership reports and social network analysis to depict the connections between shareholders and targets in a shareholder activism network. We concentrate on brokers and the 1% of target firms with the largest number of shareholder activists. In our model, brokers are nodes that connect shareholders and targets that are not directly connected. We find no evidence of herding behaviour in either brokers or the 1% of target firms. Finally, we examine the degree to which network and accounting variables contribute to the cross-sectional variation in market response to shareholder activism. To this end we apply regression analysis to detect the relationship between abnormal returns during different periods around the event of activism and various network and accounting variables. We find that liquidity related variables are statistically significant before and after the activism event. Moreover, degree and closeness are statistically significant and positive before the activism event. Regression analysis is also adopted to examine the degree that network and accounting variables affect the probability of a target firm to be targeted. We find that cash-on-hand and size related variables are statistically significant before and after the activism events as well as network variables related to the small-world phenomenon are statistically significant and positive before the activism events. Our findings are important to regulators and investors affected by principal–agent conflicts in targeted firms.
Degree
thesis:*- Grantor dc:publisher
- University of the Aegean
- Year dc:date
- 2022
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Kostaris, Konstantinos
- Κωσταρής, Κωνσταντίνος
Subjects
dc:subject × 10Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/51556