EcoSta 2023: Start Registration
View Submission - EcoSta2023
A0296
Title: Some first results from an agent-based model of consumer demand Authors:  Georgios Alkis Tsiatsios - National and Kapodistrian University of Athens (Greece) [presenting]
Iraklis Kollias - National and Kapodistrian University of Athens (Greece)
Evangelos Melas - National and Kapodistrian University of Athens (Greece)
John Leventides - National and Kapodistrian University of Athens (Greece)
Costas Poulios - National and Kapodistrian University of Athens (Greece)
Abstract: An agent-based model of consumer choice is studied. One of the primary advantages of this type of modelling is that it allows for various forms of heterogeneity between agents and different behavioural rules regarding decision-making. Here, agents are heterogeneous concerning their preferences and income. The behavioural rule is utility maximization. Commodity prices are subject to different states of the world. Hence, the model is conducive to the presence of risky outcomes. Because agent-based models are based in large part on simulations, the modeller can derive relations of a dynamic sort that are otherwise unobtainable within the framework of the representative agent model of consumer choice and classical demand theory. An example of this is that Walrasian demand now is not simply a point in commodity space but rather a complete trajectory for each agent or each group of agents. The agent-based data-driven model is used to derive demand functions of a heterogeneous set of consumers of a market consisting of different types of products. The demand functions derived are multiparametric. Graphically, various forms are depicted, conclusions are drawn about the market, and the economy is simulated.