Predictive Methods in Economics: The Link Between Econophysics and Artificial Intelligence

Abstract

In this chapter I investigate the processes and the results of quantitative methods applied to finance, giving a broad overview of the most used techniques of Complex Systems and Artificial Intelligence. Econophysics introduced in the mathematical modelling of financial markets methods such as Chaos Theory, Quantum Mechanics or Statistical Mechanics, trying to represent the behaviour of systems with a huge number of particles, while identifying human traders with particles. These models are very useful to describe and predict financial markets, especially while embedded with algorithms from Machine Learning, overcoming traditional methods from Artificial Intelligence that fail on deeply mapping the historical series on their own. The creation of structures that are not perfect on the input data but have a good accuracy on blind data becomes more and more meaningful, using sophisticated techniques of Artificial Intelligence to avoid overfitting. The combination of Artificial Intelligence and Econophysics is the key to describe complex dynamics of economic and financial world, as revealed by quant funds, constantly over benchmark, but it is of primary importance to test these innovative approaches during times of crisis such as the 2008 great recession or the 2020 pandemic.