Network theory has provided novel concepts and analytical tools for understanding how actors can leverage privileged access to others' expertise to make sound decisions. But to date it has focused on the social ties that comprise social communities. By identifying communities through network patterns of attention, instead of through patterns of direct soical network linkages, this project will identify cognitive communities, thereby reorienting network analysis from its traditional focus on social ties. To do so, a dataset will be constructed on the stock market coverage of more than 19,000 analysts for more than 16,000 companies listed on the major US stock exchange markets, during the period 1983-2011.
Intial hypotheses will be examined on how members of a relatively closed cognitive community performs in solving complex but local problems, how actors spanning structural holes perform solving less complex but more distant problems, and how actors occupying a structural fold of overlapping cohesive communities peform in solveing the most complex distant problems.
This work will shed light on how attention allocation in markets affects cognition and how the social distribution of attention gives an edge to some actors. Because the reports of analysts are a key ingredient in the trading decisions of market actors, this analysis will provide new lenses to understand how abnormal fluctuations in market prices originate, and ultimately lead to financial bubbles.
