prikaz prve stranice dokumenta Analiza vremenskih nizova zasnovana na kompleksnim mrežama
No public access
master's thesis
Analiza vremenskih nizova zasnovana na kompleksnim mrežama
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2017. urn:nbn:hr:168:520491
Mrčela, Lovre
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Mrčela, L. (2017). Analiza vremenskih nizova zasnovana na kompleksnim mrežama (Master's thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:520491

Mrčela, Lovre. "Analiza vremenskih nizova zasnovana na kompleksnim mrežama." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2017. https://urn.nsk.hr/urn:nbn:hr:168:520491

Mrčela, Lovre. "Analiza vremenskih nizova zasnovana na kompleksnim mrežama." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2017. https://urn.nsk.hr/urn:nbn:hr:168:520491

Mrčela, L. (2017). 'Analiza vremenskih nizova zasnovana na kompleksnim mrežama', Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 01 June 2024, https://urn.nsk.hr/urn:nbn:hr:168:520491

Mrčela L. Analiza vremenskih nizova zasnovana na kompleksnim mrežama [Master's thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2017 [cited 2024 June 01] Available at: https://urn.nsk.hr/urn:nbn:hr:168:520491

L. Mrčela, "Analiza vremenskih nizova zasnovana na kompleksnim mrežama", Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2017. Available at: https://urn.nsk.hr/urn:nbn:hr:168:520491

Metadata
TitleAnaliza vremenskih nizova zasnovana na kompleksnim mrežama
Title (english)Complex network based time series analysis
AuthorLovre Mrčela
MentorZvonko Kostanjčar (mentor)
Committee memberZvonko Kostanjčar (predsjednik povjerenstva)
Committee memberDražen Jurišić (član povjerenstva)
Committee memberTomislav Petković (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2017-07-18, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
U ovom radu istražene su metode za analizu financijskih vremenskih nizova, temeljene na algoritmu statističke arbitraže. Rezultati istraživanja zaokruženi su u samostalan algoritam za simulaciju trgovanja vrijednosnicama temeljenu na povijesnim podacima, i trgovanje vrijednosnicama nad podacima koji dolaze u stvarnom vremenu. Ulazne podatke za algoritam čini kretanje cijena nekog broja vrijednosnica u nekom razdoblju. Na temelju njih, konstruira se kompleksna mreža koja opisuje odnos vrijednosnica u svakom vremenskom koraku. Iz dobivene mreže procjenjuje se poredak vrijednosnica po preferenciji i računa se pouzdanost same procjene poretka. Konačno, na temelju poretka i pouzdanosti njegove procjene konstruira se portfelj koji čine vrijednosnice s najvećom preferencijom. Algoritam je simuliran nad stvarnim povijesnim podacima i polučio je uspješne rezultate. Iako su istraženi specijalno financijski vremenski nizovi, korišteni matematički i statistički postupci koji su korišteni lako mogu poslužiti i u druge svrhe.
Abstract (english)
In this thesis, methods for financial time series analysis using statistical arbitrage algorithm are researched. Results of the research are summed up into standalone algorithm for simulation of assets trading based on historical data, and also for real world assets trading on live data. Input data for the algorithm are prices of some number of assets during some time span. Next, a complex network is constructed using the input data that describes relations among the assets at the each time step. Then, the ranking based on preference of assets is derived from the network and also the confidence of derived ranking is obtained. Finally, portfolio is constructed from the most preferred assets, by using ranking and its confidence. Simulating the algorithm on the real historical data has yielded significant results. Although the research in this thesis has been done on financial time series, mathematical and statistical methods used can easily prove useful in other applications.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:520491
Study programmeTitle: 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 resourceText
File originBorn digital
Access conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2019-03-13 23:16:49