Model kompleksnog sustava zasnovan na agentima

master's thesis

• Title (english): An agent based model of a complex system
• Author: Jasmina Kolonić
• Mentor: Branko Jeren (mentor)
• Committee member: Branko Jeren (predsjednik povjerenstva)
• Committee member: Dražen Jurišić (član povjerenstva)
• Committee member: Zvonko Kostanjčar (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2014-07-08, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Tržište dionica predstavlja kompleksan sustav s velikim brojem sudionika gdje je konačno kretanje cijena dionica određeno agregiranim ponašanjem sudionika (agenata). U ovom radu predstavljen je model tržišta zasnovan na agentima. Svaki agent na temelju financijskih pokazatelja iz prošlosti predviđa buduće stanje i trguje na temelju svojih evaluacija. Izlazna cijena modela određena je trgovanjem agenata, a njihove evaluacije tvore intervale koji predstavljaju predikciju cijene. Model je testiran i uspoređen s podacima sa Zagrebačke burze. Rezultati ukazuju da model prati
kretanje cijene dionica, te pokazuju prediktivnu sposobnost za nekoliko idućih dana trgovanja.

Abstract (english)

Stock market represents a complex system with large number of participants. The prices of stocks are a result of the participants' (agents') actions. In this thesis an agent based model is given for the stock market. Each agent predicts the future price of a stock from the financial indicators from the past and trades based upon that predictions. The output price of the model is defined by the trading and agents' predictions form intervals which represent the final prediction of the price. The model was tested and compared to the data obtained from Zagreb Stock Exchange. The results show that the model follows the trajectory of the price and shows predictive power for the next couple of days.

• Keywords: Kompleksni sustavi; modeliranje zasnovano na agentima; tržište dionica; slučajne šume
• Keywords (english): Complex systems; agent based modelling; stock market; random forest algorithm
• Language: croatian
• URN:NBN: urn:nbn:hr:168:933585
• Study programme: Title: Information and Communication Technology; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka informacijske i komunikacijske tehnologije (magistar/magistra inženjer/inženjerka informacijske i komunikacijske tehnologije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2018-02-06 09:44:00