Three essays on agent-based modeling of economic networks
|Author:||PhDr. Natálie Švarcová, Ph.D. (30.9.2009)|
|Year:||2009 - summer|
|Leaders:|| † prof. RNDr. Ing. František Turnovec CSc.
|Work type:|| Dissertations
|Awards and prizes:|
|Abstract:||The thesis focuses on the question what are the economic consequences of the existence of different topologies of the economic networks on the functioning of the economic system. The thesis consists of three essays.
First essay deals with one of most fundamental questions for the functioning of markets and economic systems: emergence and persistence of cooperation among economic agents. We extend existing literature on the cooperative behavior when interactions among economic agents are structured (and modeled as networks). Compared to existing studies we simultaneously investigate the impact of the network topology and the mechanism through which particular (cooperative or defective) behavior spreads (in our case imitation strategy). Our main contribution is explicit discussion of the imitation rules based on empirically observed heuristics used by participants of behavioral experiments of Narduzzo, Warglien (1996). Agent-based computational experiments show that both topological features of particular interaction structure and evolutionary mechanism used in the model strongly influence the dynamics of the game and fraction of cooperating agents in the system.
In the second essay model of migration between national economies is introduced. Based on widely accepted Maslow’s theory of motivation decision-making function in our model includes not only wage maximization (traditionally assumed by economic models of migration) but also people’s desire to fulfill their safety and social needs. Individuals in the model are heterogeneous in the sense of different social relations they are involved in and their decision-making about migration is based on endogenously evolving network of friends. We show that inclusion of non-economic factors into decision-making of possible migrants reduces migration flows between economies for regions with developed social protection and social safety nets.
Last part of the thesis presents model of the diffusion of innovations. The results indicate that the topology of the relations among individuals importantly influences speed and development of the diffusion process as well as final market penetration. Scale free topology seems to promote fast innovation diffusion being at the same time characterized by high uncertainty of the diffusion outcomes. Less heterogeneous networks (small worlds, two dimensional lattice and ring) yield much slower diffusion of the innovation being at the same time much less unpredictable than scale free topology.
|Downloadable:|| Dissertation - N. Švarcová