Model visokofrekventnih financijskih vremenskih nizova zasnovan na strojnom učenju

master's thesis

• Title (english): Machine Learning Based Model for High-frequency Financial Time Series
• Author: Jure Rezić
• Mentor: Zvonko Kostanjčar (mentor)
• Committee member: Zvonko Kostanjčar (predsjednik povjerenstva)
• Committee member: Tomislav Petković (član povjerenstva)
• Committee member: Juraj Petrović (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-07-16, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U sklopu ovog diplomskog rada razvijen je model zasnovan na strojnom učenju financijskih vremenskih nizova dobiven iz visokofrekventnih podataka o trgovanju. Iz sirovih podataka s burze Bitstamp rekonstruirani su vremenski nizovi različitih frekvencija. U tim vremenskim nizovima izračunate su značajke koje su služile kao ulaz u model, a predstavljaju dinamiku i statistička svojstva promatranih nizova. Razvijeni model korišten je prvo za predikciju volatilnosti, koja omogućava predviđanje rizika i promjenu smjera kretanja, a zatim za predviđanje iznosa povrata na danom vremenskom intervalu.

Abstract (english)

As part of this thesis, a machine learning based model of high-frequency financial time series obtained from high-frequency trading data was developed. Time series of different frequencies were reconstructed using raw data from the Bitstamp exchange. In these time series, the features that served as an input to the machine learning model were calculated, and they represent the dynamics and statistical properties of the observed series. The developed model was used first to predict volatility, which allows to predict risk and change of the direction of price movement, and then to predict the amount of return on a given time interval.

• Keywords: strojno učenje; model; visokofrekventni financijski vremenski nizovi; statistika; financije; autokorelacija; volatilnost; povrat; regresija; klasifikacija; neuronske mreže; Bitcoin; kriptovaluta
• Keywords (english): machine learning; model; high-frequency financial time series; statistics; finance; autocorrelation; volatility; return; regression; classification; neural networks; Bitcoin; cryptocurrency
• Language: croatian
• URN:NBN: urn:nbn:hr:168:248358
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-13 13:42:43