Strong-hand conjecture: agent-based numerical simulation

Marek Karaś and Anna Serwatka

Abstract

Following the example of the Kim–Markowitz model, this study adopts similar mechanisms of market operation to perform computer simulations based on agent modeling on the financial market, where shares of one company and a bank account are available (agents can invest and borrow money). The aim of the paper is to observe how the total portfolio changes in two groups of investors: so-called large and small investors. We examine the ratio of the total portfolio of large investors over time and hypothesize that, looking at two groups of investors who are equally rich at time t = 0, the group of large investors will gain more by the last day of our simulations. Our simulations show that large investors have a greater influence on stock prices and are stronger as a group, with a total portfolio greater than that of small investors on the last day of simulations, confirming the conjecture of strong hands.