Dresden 2006 – scientific programme

Parts | Days | Selection | Search | Downloads | Help

AKSOE: Physik sozio-ökonomischer Systeme

AKSOE 12: Economic Models and Evolutionary Game Theory II

AKSOE 12.1: Talk

Thursday, March 30, 2006, 10:15–10:45, BAR 205

A Network Model of Company Growth — •Michael Koenig, Stefano Battiston, and Frank Schweitzer — Chair of Systems Design, ETH Zurich,CH-8092 Zurich,Switzerland

We study a network model of company growth in which each node represents a company. Growth of a company is described by means of a differential equation that includes a stochastic auto-catalytic term as well as terms for the interaction with other companies. These interactions contribute positively to growth and can be interpreted as the sharing of knowledge. Global couplings stand for limited resources and external conditions, e. g. market restrictions imposed by governmental institutions.

We test several local rules of network formation which specify how companies choose other companies to cooperate with. All these rules assume selfishness and bounded rationality, the latter meaning that each company has only local information on the network topology and makes decision on a trial and error basis.

By means of computer simulations we investigate the conditions for the emergence of non-random structures. We observe that some specific rules lead to the appearance of alliances of cooperating companies, the formation of hubs (companies that are highly connected) and structures with a hierarchy of size levels. We find that during the evolution of the network starting from a random graph, the clustering coefficients and other quantities related to the length of cycles display abrupt changes reflecting the appearance of hierarchical structures.