Procjena kretanja cijena dionica neuronskim mrežama

undergraduate thesis

• Title (english): Estimating the share price movement with neural networks
• Author: Tomislav Kožul
• Mentor: Marko Đurasević (mentor)
• Committee member: Marko Đurasević (predsjednik povjerenstva)
• Committee member: Nikolina Frid (član povjerenstva)
• Committee member: Josip Šarić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-07-06, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Rad obuhvaća izradu modela neuronske mreže koja koristi ćelije s dugoročnom memorijom (LSTM ćelije) za pronalaženje pravilnosti i uzoraka u kretanju cijena dionica. Dobiveni model implementiran je u skriptu koja, nakon što korisnik unese simbol dionice ili fonda čiju cijenu želi predvidjeti, preuzima podatke o aktualnim cijenama, obrađuje ih te predikcije prikazuje u numeričkom i grafičkom obliku. S ciljem pronalaska modela čije predikcije imaju najmanja odstupanja od stvarnih cijena, ispitane su različite arhitekture neuronskih mreža, kao i različite duljine ulaznih podataka. Dok je model učen i ispitan isključivo na povijesnim podacima, skripta model koristi za rad s tekućim vrijednostima.

Abstract (english)

The paper includes the construction of a neural network model that uses cells with long short-term memory (LSTM cells) to find regularities and patterns in stock price movements. The obtained model is implemented into a script that, after the user enters the symbol of the stock or fund whose price he wants to predict, downloads data on current prices, processes it, and displays the predictions in numerical and graphical format. To find a model whose predictions have the smallest deviations from the actual values, different architectures of neural networks were tested, as well as different lengths of input data. While the model was trained and tested exclusively on historical data, the script uses the model to work with current values.

• Keywords: predikcija cijena dionica; LSTM; neuronske mreže
• Keywords (english): stock price prediction; LSTM; neural networks
• Language: croatian
• URN:NBN: urn:nbn:hr:168:189510
• 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: 2023-10-11 09:23:27