Agent-Based Simulation in Finance

Mutli-agent based simulation models allow for the research of economic questions differently than before. This thesis builds a multi-agent based simulation model of the U.S. financial market, to examine the behavior of banks within it. Although the focus is on the credit market, a sub-goal is to build a complete model of the market. This is due to the observation, that in the 2007 financial crisis, seemingly small adjustments in one market of the U.S. led to a world-wide shock to the financial system. While designing, some existing ACE-models are adapted as well as new ones incorporated. The simulation model includes ideas for models of banks, hedge funds, the Federal Reserve System, a stock trading simulation, an information market and federal, as well as corporate and private household, credit customers.

Autorentext

Jason Thomas Wilmans was born 1985 in Kiel. After finishing high school he initially spent a few years working in the social sector. He started studying applied computer sciences at the University of Applied Sciences Hamburg in 2010. Since starting his master studies he specializes in distributed, large scale, agent-based, socio-economic models.