Title: A General Model of Socio-Economic Right-Skew Distributions

Speaker: Paul Ormerod, Volterra Consulting www.volterra.co.uk       

Date/Time: Monday, December 1, 2008, 3:00 – 4:00 pm

Location: CSRI Building/Room 90 (Sandia-NM)

Brief Abstract: We identify a clear relationship which is observed empirically in social, economic and cultural settings between the power law exponent and the fraction of the whole data which is characterized by the power law tail. We present a parsimonious agent-based model, with behavioural underpinnings, which is able to replicate this relationship. In addition, the model is able to generate a wide range of size-rank distributional outcomes, such as power law across the whole data, power law in the tail, winner take all and other right-skew distributions which are less easy to classify in a standard way. A special form of the model is already known in physics and genetics. But we generalize from this, and the generalization is essential to the rich variety of distributional outcomes which the model can generate. Finally, the model is able to account for the continuous turnover of the rankings of individual agents within any given overall size-rank distribution which is observed in most cultural, social and economic settings.

CSRI POC: Randall Laviolette, (505) 284-1325