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

Figure 8

Average relative percentage gain of wealth of the 100 agents for each type of network. Values are ranked from 1 (highest gain) to 100 (greatest loss). The way agents are connected in the Barabási network favors a dynamic of strategy change. On average, most agents are richer than if they did not change their strategy, with a minimum number of agents with almost zero losses. The random network has similar behavior with slightly lower gain values than the Barabási network for most agents. The small-world network has lower values of wealth gain than the other two networks and higher values of wealth loss for about 30% of investors compared to the other networks.