Abstract
A financial agent-based price model is developed by a combination of the contact system and the social network theory. Agents are supposed to live on a small-world network described by the Watts–Strogatz model and their attitude interinfluences along the network are characterized by particle interactions in the Markov contact process. The financial time series of the model are generated by Monte Carlo simulations and analysed with respect to a number of “stylized facts”, focusing on the role played by network topology. Return series of the model can reproduce nonnormality, volatility clustering and multifractality, consistent with an empirical observation of real market data. And these “stylized facts” are most salient when the model is of the small-world network topology rather than completely ordered or random one in our experiments, highlighting a potential influence of the structure of agent attitude exchange channel on financial price fluctuation behaviours.
Published Version
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