The Methodology of Agent Based Modelling Motivation and

The Methodology of Agent. Based Modelling - Motivation and Applicability Charlotte Bruun

Introductory examples • Boids: http: //www. red 3 d. com/cwr/boids/ • Game of life: http: //www. bitstorm. org/gameoflife/ – glider gunn http: //en. wikipedia. org/wiki/Conway%27 s_Game_of_Life • Sugarscape http: //www. brookings. edu/es/dynamics/sugarscape/default. htm • Schellings seggregation model http: //www. econ. iastate. edu/tesfatsi/demos/schelling/schellhp. htm • http: //www. wayner. org/texts/seg/012298 segregate-sim 2. html • Artificial life

Complex Adaptive Systems • Large number of interacting agents • Roughly, by a complex system I mean one made up of a large number of parts that interact in a non simple way. In such a system the whole is more that the sum of its parts, not in a ultimate metaphysical sense but in the important pragmatic sense that, given the properties of the parts and the laws of their interaction, it is not a trivial matter to infer the properties of the whole. Simon (1981) • Time AND space – Ordinary differential equations: no space, partial: continous space • Selforganization – swarms of insects • Emergence – properties of the whole

Agent-Based Modelling • • • A method for understanding complex dynamic systems. Bottom-up rather than top-down Local interaction in space and time No central coordination (insect ”queen”) Elements: – Environment (on which agents operate, with which they interact) – Agents (internal states, behavioural rules) (heterogeneity) • • Autonomy Social ability Reactivity Proactivity – Rules (agent-environment, agent-agent, environment-environment) • Vary parametrically, not structurally

Methodological individualism? • • • Non-reductionistic approach Social phenomena Neither Induction or deduction Synthetic approach We can only understand (social) phenomena by growing them! • More than one agent – interaction • Rationality with the system rather than with the agent • Dumb agents in a complex world rather than rational agents in a simple world.

Transaction Cost Economics • Search and information costs are costs such as those incurred in determining that the required good is available on the market, who has the lowest price, etc. • Bargaining costs are the costs required to come to an acceptable agreement with the other party to the transaction, drawing up an appropriate contract and so on. • Policing and enforcement costs are the costs of making sure the other party sticks to the terms of the contract, and taking appropriate action (often through the legal system) if this turns out not to be the case. • How can boundedly rational agents, acting in time and space, minimize transaction costs?

Institutions • Systems of established and prevalent social rules (language, money, law ect. ) (Hodgson 2006) • Institutions constrain and enable behaviour • Institutions may be emergent phenomena – Repeated prisonners dilemma: http: //en. wikipedia. org/wiki/Prisoner's_dilemma – Coordination game: http: //en. wikipedia. org/wiki/Coordination_game – Emergence of money • Hodgson: In a world of incomplete and imperfect information, high transaction costs [. . etc] powerful institutions are necessary to enforce rights. (can only the state do this? ) • Agent-based models may take institutions as given, or try to capture their emergence.

Path dependence – positive feedback • Qwerty • Positive feedback as a way of solving social problems – eg. Money • Path dependence not optimality calculations • Solutions may not be optimal – but they work • No search for equilibria

Scientific Explanation • What constitutes an explanation of an observed social phenomenon? – % can you explain it? – + can you grow it? – What set of microspecifications are sufficient to generate the macrophenomena? – A generative social science