Přístupnostní navigace
E-application
Search Search Close
Master's Thesis
Author of thesis: Ing. Erik Maňásek
Acad. year: 2025/2026
Supervisor: Ing. Pavel Hrabec, Ph.D.
Reviewer: doc. Mgr. Zuzana Hübnerová, Ph.D.
This thesis presents a~systematic comparison of autoregressive models for financial time series forecasting. Four mean models -- SARIMA, ARFIMA, ETS and TBATS -- are evaluated in combination with GARCH and EGARCH volatility models and four innovation distributions. The models are assessed on 667 daily series from the US stock market using a~rolling one-step-ahead forecast over 2025, employing point, interval, volatility and probabilistic forecast metrics. Results show that advanced models significantly outperform the naive random walk benchmark in approximately 10--13\,\% of cases after applying diagnostic filtering conditions, with TBATS dominating in probabilistic metrics and ETS in point accuracy. The thesis outlines how these results can be used to improve inputs for portfolio optimization beyond the Markowitz framework with naive forecasts.
time series, SARIMA, ARFIMA, ETS, TBATS, GARCH, EGARCH, rolling forecast, forecast metrics
Date of defence
08.06.2026
Result of the defence
Defended (thesis was successfully defended)
Grading
A
Process of defence
Student odprezentoval svoji práci. Byly přečteny posudky vedoucího a oponentky. Vedoucí nebyl přítomen. Oponentka položila studentovi otázky z posudku, na které student odpověděl.
Language of thesis
Czech
Faculty
Fakulta strojního inženýrství
Department
Institute of Mathematics
Study programme
Mathematical Engineering (N-MAI-P)
Composition of Committee
prof. RNDr. Zdeněk Pospíšil, Dr. (předseda) prof. Mgr. Pavel Řehák, Ph.D. (místopředseda) doc. Mgr. Zuzana Hübnerová, Ph.D. (člen) doc. Mgr. Zdeněk Opluštil, Ph.D. (člen) doc. Mgr. Jaroslav Hrdina, Ph.D. (člen)
Supervisor’s reportIng. Pavel Hrabec, Ph.D.
Grade proposed by supervisor: A
Reviewer’s reportdoc. Mgr. Zuzana Hübnerová, Ph.D.
Grade proposed by reviewer: A
Responsibility: Mgr. et Mgr. Hana Odstrčilová