Abstract
This paper demonstrates the multilayer structure of information in financial markets. While only 3.59% of 8,190 stock/quarter pairs have single information layer, 75% have two to five layers and 18% have six to eight layers. We develop a clustering algorithm which determines the number of information layers with the accuracy rates between 95% and 97% in simulated datasets. We propose a multilayer informed trading measure, MPIN by extending the original model of Easley et al. (1996). Respective solutions for computational problems regarding MPIN estimation are also provided. PIN model's estimates on informed trading and five intermediate parameters are heavily biased for datasets with multiple information layers, while MPIN model consistently provides accurate estimates. When compared to PIN, MPIN yields substantially higher probabilities of informed trading and information event occurrence; much lower rate of informed traders for real stock/quarter pairs. We obtain daily posterior multilayer probabilities conditional on the quarterly estimates of MPIN model. Posterior probabilities derived from MPIN model are approximately two fold when compared to the ones from PIN model surrounding earnings announcements.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
