It is a matter of fact that a majority of people are affected directly or indirectly by stock market activity. They are affected directly, by trading assets in stock markets, and indirectly when savings are traded by institutions in the stock market. The aim of this project is to investigate the dynamics of the trading done in stock markets. To achieve this aim, you will be involved in the development of a computer model that tunes between the Brownian model of noisy traders and the neoclassical model where the trading agents are fully informed about the market and they make decisions by maximizing their utility.
The stock market is represented as a system of heterogeneous agents with bounded rationality and limited information. With the purpose to achieve the most accurate model, two well-defined methodologies will be developed: (1) An agent-based algorithm to model stock markets where noisy players, fundamentalists, and strategists trade under the effect of confidence, price/value ratio, and herding interaction; and (2) a web-based game to mimic stock markets using human participants. The web-game is accessible by an app where users can enter into the game and take actions (buy, sell, or hold stocks) according to the price fluctuations, news, and rumors spread through a self-regulated agent network. This platform will be used as an experimental framework to validate our algorithm, and it will provide experimental data on individual agent’s behavior that will help to validate the model and to better understand stock markets.
You need deep knowledge in programming, mathematics, statistical, and finance.
The opportunity ID for this research opportunity is 2176