Institutional Repository of Vilnius University
Kriptovaliutų kainų prognozavimas laiko eilutėse /
Abstract
dc:descriptionStock price forecasting has always been an important topic in the area of ​​finance and economics. Although stock price forecasting is known to be one of the most challenging financial tasks due to the dynamic and unpredictable nature of these markets, many researchers are driven to develop forecasting models that would not only be reliable and accurate, but would also work for all types of instruments. This thesis aims to explore different types of forecasting autoregressive models such as ARMA, ARIMA and ARIMA-GARCH for cryptocurrency prices based on its previous time series and understand if such autoregressive models are potentially accurate and could be used to predict future values ​​for cryptocurrency prices. Therefore, to find a suitable model to get now accurate predictions several calculations have been done firstly to prepare the data for predictions and secondly to detect the best model parameters. For model reliability, actual prices and predicted prices were compared and were graphically plotted in the diagrams. During this experiment, different range of prices were being predicted, however, only the smallest range of prices returned best results and could not predict accurate values ​​for a longer period. In conclusion, after comparing the results, no such autoregressive model was found to be used as a reliable tool for cryptocurrency prices due to its dynamic and volatility.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2022
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Sartauskaitė, Aušrinė,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD146237086&prefLang=en_US