Regensburg 2013 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
SOE: Fachverband Physik sozio-ökonomischer Systeme
SOE 4: Economic Growth and Longevity I
SOE 4.3: Talk
Monday, March 11, 2013, 12:00–12:15, H37
Statistical mechanics of organizational growth processes — •Hernan Mondani1, Petter Holme2, and Fredrik Liljeros1 — 1Department of Sociology, Stockholm University, 106 91 Stockholm, Sweden — 2Department of Physics, Umeå University, 901 87 Umeå, Sweden
In this study we address the question of the origin of non-Gaussian, fat-tailed growth-rate distributions in organizational growth processes.
In particular, the remarkable feature that the probability distribution for the growth rate –i.e. how quickly the size changes –follows a non-Gaussian pattern, something unexpected in traditional approaches. In practice, this means that organizational size changes very little most of the time, but dramatically every once in a while.
This pattern shows up in growth processes of various unrelated systems, from bird migration dynamics to investments in mutual funds.
We verify the emergence of this pattern on real data from voluntary organizations in Sweden.
We implement a stochastic model on a contact network with influence-based time evolution. We simulate the model for different network configurations, and are able to reproduce the key features with this model.
We then construct a physical analogy to the model, motivated by the sociological concept of preference falsification. This allows us to find two distinct phases emerging in extreme cases of the control parameter.
Keywords: sociophysics, organizational growth process, tent-shaped distribution, complex networks, preference falsification.