prikaz prve stranice dokumenta Predikcija cijena dionica umjetnom inteligencijom
No public access
master's thesis
Predikcija cijena dionica umjetnom inteligencijom
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2021. urn:nbn:hr:168:374452
Baković, Krešimir
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Baković, K. (2021). Predikcija cijena dionica umjetnom inteligencijom (Master's thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:374452

Baković, Krešimir. "Predikcija cijena dionica umjetnom inteligencijom." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2021. https://urn.nsk.hr/urn:nbn:hr:168:374452

Baković, Krešimir. "Predikcija cijena dionica umjetnom inteligencijom." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2021. https://urn.nsk.hr/urn:nbn:hr:168:374452

Baković, K. (2021). 'Predikcija cijena dionica umjetnom inteligencijom', Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 12 July 2024, https://urn.nsk.hr/urn:nbn:hr:168:374452

Baković K. Predikcija cijena dionica umjetnom inteligencijom [Master's thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2021 [cited 2024 July 12] Available at: https://urn.nsk.hr/urn:nbn:hr:168:374452

K. Baković, "Predikcija cijena dionica umjetnom inteligencijom", Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2021. Available at: https://urn.nsk.hr/urn:nbn:hr:168:374452

Metadata
TitlePredikcija cijena dionica umjetnom inteligencijom
Title (english)Prediction of stock prices by artificial intelligence
AuthorKrešimir Baković
MentorVinko Lešić (mentor)
Committee memberVinko Lešić (predsjednik povjerenstva)
Committee memberMato Baotić (član povjerenstva)
Committee memberBranimir Novoselnik (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2021-07-06, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Electrical Engineering
Abstract
Tržište vrijednosnih papira je jedno od najpromatranijih te najvećih tržišta općenito te kao takvo privlači brojne investitore. Upravo navedeno omogućava moguće ostvarivanje profita tj. zaradu. S ciljem predikcije kretanja cijene dionice, u ovom diplomskom radu izrađena su 2 modela koji vrše predikciju za 1 dan unaprijed. Programska podrška za modele izvedena je kao Python skripta unutar virtualnog okruženja PyCharm. Oba modela vrše predikciju zaključne cijene dionica američkih tvrtki Tesla i Twitter na temelju prošlih zaključnih cijena, prodajnog volumena te derivacije cijene dionice. Kao korišteni modeli koristili su se GRU i LGBM model te su oba modela dali zadovoljavajuće rezultate koji govore kako bi se modeli mogli koristiti kao pomoćni alat pri trgovanju. Srednja apsolutna greška za GRU iznosi 1.2325 dok je ona za LGBM 1.3689.
Abstract (english)
The financial market is one of the most observed and most important markets in general and as such attracts many investors. Precisely the above enables the possible realization of profit. In order to improve the prediction in the movement of the stock price, in this paper 2 models were made that perform prediction for 1 day in advance. Model software is implemented as a Python script within the PyCharm virtual environment. Both models predict the closing price of stocks of the American companies Tesla and Twitter based on past closing prices, sales volume and derivation of the stock price. The GRU and LGBM models were used as models and both models gave satisfactory results which suggest that the models could be used as an auxiliary trading tool. The mean absolute error for GRU is 1.2325 while for LGBM is 1.3689.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:374452
Study programmeTitle: Electrical Engineering and Information Technology
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka elektrotehnike i informacijske tehnologije (magistar/magistra inženjer/inženjerka elektrotehnike i informacijske tehnologije)
Type of resourceText
File originBorn digital
Access conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2023-01-13 11:42:57