A threshold model of financial markets

TL;DR

This paper introduces a threshold model of financial markets using an Ising spin framework where agents can buy, sell, or stay inactive, successfully replicating key market phenomena like fat-tailed returns and volatility clustering.

Contribution

It presents a novel agent-based model incorporating three agent actions and a magnetization-based price evolution, capturing essential stylized facts of real markets.

Findings

Reproduces fat-tailed distribution of returns

Captures volatility clustering

Models market dynamics with Ising spin interactions

Abstract

We proposed a model of interacting market agents based on the Ising spin model. The agents can take three actions: "buy," "sell," or "stay inactive." We defined a price evolution in terms of the system magnetization. The model reproduces main stylized facts of real markets such as: fat-tailed distribution of returns and volatility clustering.