Investor Networks in the Stock Market

Abstract

We study investor networks in the stock market, through the lens of information network theory. We use a unique account level dataset of all trades on the Istanbul Stock Exchange in 2005, to identify traders who are similar in their trading behavior as linked in an empirical investor network (EIN). This empirical investor network is consistent with several predictions from the theory of information networks. The EIN is relatively stable over time, centrally placed investors earn higher profits, some investors systematically trade before their neighbors in the network, and the cross sectional distributions of profits and trading volume are heavy-tailed with similar tail exponents. We also identify interesting theoretical challenges for future research.

Keywords: Information Networks, Heterogeneous Investors, Portfolio Choice, Power Law

JEL Classification: G11, G12