Modelización del mercado financiero como un Sistema Complejo Adaptativo

Abstract

The theory of complex adaptive systems is still in development. In fact, this type of systems is currently studied by simulating specific examples. Simulating different kinds complex adaptive systems allows us to uncover some of the properties of such systems, which may contribute to the construction of a future formal theory.

In this thesis, we shall try to model some aspects of the macro economy by means of a complex adaptive system. In particular, we develop a computational model of a financial market, which includes the impact of wealth phenomenon. The model is built in two stages. In the first stage, we design the characteristics of “economic investors” by using a specific learning classifier system (which is a machine learning structure based on genetic algorithms). In the second stage, we build the corresponding market structure, which allows investors to interact to each other by buying and selling stocks. We find that our model is able to qualitatively reproduce the time series’ features of real stock prices. In addition, our model, with a few, appropriate modifications, is also able to reproduce the economic features predicted by the classic economic theory (based on the “efficient markets” hypothesis).