The Influence of Confidence and Social Networks on an Agent-Based Model of Stock Exchange

Figure 7

Investment strategy spread rate (%) according to the network topology. For the Barabási network (in red), there is an increasing propagation of strategy from the initial moment of the simulation; that is, the agents are leaving their initial strategies for others, reaching rapidly maximum values from which they decrease until reaching low values (less than 10). This indicates that a significant portion of the agents, after switching to a given strategy, return to their initial strategies. In the small-world network (in green), the maximum values (of the same order as the Barabási network) are reached over a longer time and decrease to values that are not too low, that is, the dynamics of change is such that a larger portion of agents go back to their initial strategy and change again. As for the random network (blue), its maximum value is well below that of the other networks.