Trading Signatures: Investor Attention Allocation in Stock Markets

Abstract

Investors in stock markets face a huge amount of financial information. For that reason, they must decide how to distribute their limited attention across different securities. We propose a new measure of investor attention, called the trading signature. This measure allows us to analyze whether there are heterogeneous, persistent, and characteristic patterns in the way how investors distribute their attention. Our study builds on a large data set that contains every transaction of hundreds of thousands of households and financial institutions in a stock market for over 20 years. We find that the ways how investors distribute their attention are not only distinctive but invariant with respect to changes in their portfolios. Moreover, investor behavior is surprisingly persistent: it takes at least seven years for an investor to significantly change her attention distribution. These observations are strikingly similar to recent findings on how people manage their social networks. However, in contrast to social relationships, time constraints do not appear to limit investors’ activeness in the markets: the larger is an investor’s portfolio, the more frequently each security is traded on average. A possible explanation is that different trading strategies require different amounts of effort, and an investor can switch between the strategies if she changes the size of her portfolio.

Keywords: trading signatures, complex networks, investor behavior, stock market

JEL Classification: G11