Algoritmi za particioniranje grafova zavisnosti financijskih vremenskih nizova

undergraduate thesis

• Title (english): Algorithms for Partitioning Dependence Graphs of Financial Time Series
• Author: Pavao Dužević
• Mentor: Zvonko Kostanjčar (mentor)
• Committee member: Zvonko Kostanjčar (predsjednik povjerenstva)
• Committee member: Stjepan Begušić (član povjerenstva)
• Committee member: Vanessa Keranović (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2021-09-02, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Cilj ovog rada je predstaviti metodu formiranja grafa iz matrice korelacije vremenskih nizova te prikazati njegova svojstva u svrhu particioniranja i posljedično tome izgradnje hijerarhijske strukture. Takva hijerarhijska struktura grupira slične vremenske nizove u podgrafove izvornog grafa. Koristeći opisanu metodologiju, gradimo i ispitujemo svojstva takvih grafova na vremenskim nizovima povrata financijskih vrijednosnica. Algoritam za particioniranje grafa koji se koristi u svrhu dobivanja rezultata unutar ovog rada je algoritam spektralne bisekcije. U konačnici, algoritam je ispitan na vremenskim nizovima cijena dionica Američkog indeksa 1000 najvećih kompanija po kapitalizaciji – Russell 1000.

Abstract (english)

The goal of this thesis is to present a method of graph forming from correlation matrix of several time series and to show the graph's properties with the purpose of partitioning it and, as a consequence, building a hierarchical structure. Such a hierarchical structure groups similar time series in subgraphs of the original graph. We build and test properties of said graphs on financial securities return time series by using the described methodology. The graph partitioning algorithm used to attain results within this thesis is a spectral bisection algorithm. In the end, the algorithm is tested on stock price time series of the 1000 largest companies by capitalisation, as categorised by the American index - Russel 1000.

• Keywords: teorija grafova; algoritmi particioniranja grafova; spektralna bisekcija; hijerarhijsko stablo; metode grupiranja; vremenski niz; financijske vrijednosnice
• Keywords (english): graph theory; graph partitioning algorithms; spectral bisection; hierarchical tree; clustering methods; time series; financial securities
• Language: croatian
• URN:NBN: urn:nbn:hr:168:834899
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-14 20:21:04