Fraudulent agents in an artificial financial market

TL;DR

This paper investigates fraudulent behavior in an artificial financial market using an agent-based model to analyze its impact on prices and wealth distribution, providing insights into illegal trading practices.

Contribution

It introduces a simplified model of fraudulent agents within the Genoa Artificial Stock Market to study their effects on market statistics and wealth distribution.

Findings

Fraudulent agents alter price statistical properties.

Fraudulent behavior impacts wealth distribution among agents.

The model helps identify signatures of fraud in market data.

Abstract

The problem of insider trading and other illegal practices in financial markets is an important issue in the field of financial regulatory policies. Market control bodies, such as the US SEC or the Italian CONSOB regularly perform statistical analyses on security prices in order to unveil clues of fraudulent behaviour within the market. Fraudulent behaviour is connected to the more general problem of information asymmetries, which had already been addressed in the field of experimental economics. Recently, interesting conclusions were drawn thanks to a computer-simulated market where agents had different pieces of information about the future dividend cash flow of exchanged securities. Here, by means of an agent-based artificial market: the Genoa Artificial Stock Market (GASM), the more specific problem of fraudulent behaviour in a financial market is studied. A simplified model of fraudulent behaviour is implemented and the action of fraudulent agents on the statistical properties of simulated prices and the agent wealth distribution is investigated.