This course provides an introduction to the use of Statistical Physics tools for the analysis and understanding of socio-economic systems. Most systems in this realm are composed by elements (generically termed agents) that do not act in isolation and are inhomogeneous. As a result of their interaction, complex behaviour is usually observed at a macroscopic scale and the systems operate usually in out-of-equilibrium conditions. Therefore, advanced techniques from Statistical Mechanics are suited for understanding some of their properties.
The objectives of the course is to allow the students to understand the parsimonious modelling approach taken to uncover the mechanisms behind large-scale phenomena and assimilate its underlying difficulties, limitations and strengths. Special emphasis will be given at the interpretation of the parameters included in the models. A second objective is to give the students a glimpse on state-of-the-art techniques for data analysis and model validation. All the activities of the course include training in simulation and data analysis as an integral part of the activities, for the students to gain practical experience.
Prof. Dr. Claudio J. Tessone
PhD Students of Natural Sciences (mainly Physics, Mathematics and Informatics)
Ciudad Universitaria, Universidad de Buenos Aires
Content:The course is organised in two main parts: Complex Networks and Modelling. The first part of the course will include (but be not limited to): multiplicative growth processes (which pervade the growth of technological and social networks), phase transitions in network topology (like the small-world effect, transitions from hierarchical to decentralised networks), mesoscale network properties (like communities), scaling laws (of trees) and different kinds of percolation transitions (when subjected to random or targeted attacks).
The second part of the course deals with simple models of socio-economic systems some of which were inspired in Statistical Physics, some others that exhibit a rich behaviour when analysed with such techniques. These include: Voter model (which exhibits coarsening without surface tension), Axelrod model (which exhibits transitions from homogeneous to disordered states), Threshold models (sensibility to initial conditions), and processes of diffusion and spreading (epidemiology and product adoption) which exhibit fundamentally different properties than on regular lattices.
Python and/or C, C++, Java are necessary requirement. Statistical Mechanics or Nonlinear Dynamics courses are desirable (but can be spoken about on a per case basis)
Slanina, F. (2013): Essentials of Econophysics Modelling, Oxford University Press.
Participation, group work, presentation.