Predviđanje cijena dionica pomoću analize vremenskih serija

master's thesis

• Title (english): Stock Price Prediction Using Time Series Analysis
• Author: Ivona Mramor
• Mentor: Marin Šilić (mentor)
• Committee member: Marin Šilić (predsjednik povjerenstva)
• Committee member: Klemo Vladimir (član povjerenstva)
• Committee member: Goran Delač (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2021-07-05, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U sklopu ovog rada napravljeno je predviđanje cijene dionica bazirano na metodama analize vremenske serije. Objašnjene su vremenske serije i financijske vremenske serije. Opisan je programski i arhitekturno sustav za predviđanje cijene dionica. Sustav se bazira na tri modela: model ekstremnog gradijentnog pojačanja-XGBoost, Facebook Prophet te mreža s dugom kratkoročnom memorijom-LSTM. Sva tri modela baziraju svoja predviđanja cijene dionica na cijenama dionica od prethodnih dana. Detaljno je teorijski objašnjen svaki model i prikazani koraci razvoja konačnih modela. Pojašnjeni su skupovi podataka od tri dionica koji su korišteni za ispitivanje modela. Za svaki model i za svaku dionicu detaljno su objašnjeni i uspoređeni rezultati.

Abstract (english)

In this thesis, a stock price forecast based on time series analysis methods was made. Time series and financial time series are explained. A program and architectural system for predicting stock prices is described. The system is based on three models: the model of extreme gradient boosting-XGBoost, Facebook Prophet and the network with long short-term memory-LSTM. All three models base their stock price forecasts on stock prices from previous days. Each model is theoretically explained in detail and presented the steps of development of the final model. The datasets for three stocks used to test the model are explained. For each model and for each stock, the results were explained and compared in detail.

• Keywords: Analiza vremenskih serija; predviđanje vremenskih serija; dionice; strojno učenje; LSTM; Facebook Prophet; XGBoost
• Keywords (english): Time series analysis; time series prediction; stocks; machine learning; LSTM; Facebook Prophet; XGBoost
• Language: croatian
• URN:NBN: urn:nbn:hr:168:670131
• 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 13:16:21