Javascript must be enabled for the correct page display

Simulation of information markets to determine final equilibrium price

de Boer, Caitlin (2019) Simulation of information markets to determine final equilibrium price. Bachelor's Thesis, Artificial Intelligence.

[img]
Preview
Text
AI_BA_2019_CaitlindeBoer.pdf

Download (752kB) | Preview
[img] Text
toestemming.pdf
Restricted to Registered users only

Download (139kB)

Abstract

Information markets are exchange-traded markets that are used in eliciting information about the certain likelihood of an event. With the trading of securities based on the beliefs of the traders, a final price is determined which can also be translated as the probability of an event happening. Two simulations were implemented using a simple mathematical model with different algorithms that emulate the way agents trade in such markets and different market strategies were adopted by the agents. Running the simulations with two methods (All Agent Method and Single Agent Method) over a number of iterations, we find out if the price at equilibrium was higher than that of the average belief of an agent. Other variables were tested such as the size of the unit of currency, as well as the number of agents to find out if such factors had an impact on the final equilibrium price of the market. Results showed that the higher the median belief, the higher the difference between the final equilibrium price and the average belief. The symmetry of the belief distribution also had an impact on the difference as well, whereby the skew to the left resulted in a drop in the difference while a right skew resulted in an increase in price difference. The size of the unit of currency also affected the final equilibrium price whereby a smaller unit of currency resulted in a higher equilibrium price and the number of agents had a strong effect on the Single Agent Method but not on the All Agent Method.

Item Type: Thesis (Bachelor's Thesis)
Supervisor:
Supervisor nameSupervisor E mail
Grossi, D.D.Grossi@rug.nl
Degree programme: Artificial Intelligence
Thesis type: Bachelor's Thesis
Language: English
Date Deposited: 29 Oct 2019
Last Modified: 30 Oct 2019 10:58
URI: http://fse.studenttheses.ub.rug.nl/id/eprint/21155

Actions (login required)

View Item View Item