VISUALIZING STOCK-MUTUAL FUND RELATIONSHIPS THROUGH SOCIAL NETWORK ANALYSIS
California State University, USA.
In this paper we analyze the network structure of stocks (as actors in a 2-mode/affiliation social network) and their relationship to mutual funds. The analysis reveals a network structure that has both the "hub" and "small world" characteristics of many common social and physical networks thus suggesting that stock affiliation in mutual funds is not a random phenomenon even though the mutual fund selection was done randomly. The data for the analysis is based on a random sample of 18 mutual (stock) funds from the Vanguard and Fidelity family of funds and 99 unique stocks which were part of the 10 top holdings in each fund. This study may also suggest that institutional investors are prone to herd behavior (a social network phenomena), and/or risk aversion shown by the high concentration of (similar) blue-chips in their portfolios.
Key words: Social networks, small world networks, clustering coefficient, mutual funds, herd behavior, risk aversion