Modeliranje i prognoziranje prinosa odabranih europskih burzovnih indeksa

master's thesis

• Title (english): Modeling and forecasting the returns of selected European stock indices
• Author: Klara Čelan
• Mentor: Ivana Lolić (mentor)
• Committee member: Ivana Lolić (predsjednik povjerenstva)
• Committee member: Petar Sorić (član povjerenstva)
• Committee member: Zrinka Orlović (član povjerenstva)
• Granter: University of Zagreb; Faculty of Economics and Business; (Department of Managerial Economics); Zagreb
• Defense date and country: 2024-07-05, Croatia
• Scientific / art field, discipline and subdiscipline: SOCIAL SCIENCES; Economics; Business Mathematics and Statistics

Abstract

Razvoj financijske ekonometrije i prediktivne analitike omogućio je da se na financijskim
tržištima koje karakterizira slabi oblik hipoteze o efikasnosti tržišta prognoziraju buduće cijene
i prinosi s obzirom na zabilježene povijesne podatke. Predmet je ovog rada modeliranje i
kratkoročno prognoziranje prinosa pet odabranih europskih dioničkih burzovnih indeksa (DAX,
FTSE 100, CAC 40, IBEX 35, Euro Stoxx 50) korištenjem naivnih modela, modela jednostavnih
pomičnih prosjeka, autoregresijskih modela i modela geometrijskog Brownovog gibanja s
ciljem ocjene njihove prediktivne moći. Empirijska je analiza provedena na javno dostupnim
podatcima s platforme Yahoo Finance korištenjem programskog jezika Python. Inicijalna
primjena modela dodatno je evaluirana dijeljenjem podataka na poduzorke za treniranje i
testiranje modela. Grafička i kvantitativna komparativna analiza pomoću RMSE pokazatelja
potvrđuju mogućnost uspješnog kratkoročnog prognoziranja dnevnih prinosa relevantnim
modelima.

Abstract (english)

The development of financial econometrics and predictive analytics has enabled forecasting
future prices and returns based on histrical data in financial markets that imply weak form of
the efficient market hypothesis. This paper focuses on modeling and short-term forecasting
returns of five selected European stock market indices (DAX, FTSE 100, CAC 40, IBEX 35, Euro
Stoxx 50) using naïve models, simple moving average models, autoregressive models, and
geometric Brownian motion in order to evaluate their predictive power. The empirical analysis
was conducted using publicly available data from Yahoo Finance, utilizing Python
programming language. The initial application of the models was further evaluated by splitting
the data into training and testing subsets. Graphical and quantitative comparative analysis
using the RMSE indicator confirms the possibility of successful short-term forecasting of daily
returns by applying relevant models.

• Keywords: prognoziranje; naivni modeli; modeli jednostavnih pomičnih prosjeka; AR modeli; geometrijsko Brownovo gibanje; RMSE
• Keywords (english): forecasting; naive models; simple moving average models; AR models; geometric Brownian motion; RMSE
• Language: croatian
• URN:NBN: urn:nbn:hr:148:868133
• Study programme: Title: Business economics; Study programme type: university; Study level: integrated undergraduate and graduate; Academic / professional title: sveučilišni magistar ekonomije/sveučilišna magistra ekonomije (sveučilišni magistar ekonomije/sveučilišna magistra ekonomije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2024-07-07 19:33:20