Mehrhorizont-Prognose der Bitcoin-Renditen mit ARMA-GARCH- und neuronalen Ergänzungsmodellen (2014–2024)

Multi-Horizon Forecasting of Bitcoin Returns Using ARMA-GARCH and Neural-Augmented Models (2014-2024)

Sarah Goldman ()
Additional contact information
Sarah Goldman: Lux-SIR, CRIISEA - Centre de Recherche sur les Institutions, l'Industrie et les Systèmes Économiques d'Amiens - UR UPJV 3908 - UPJV - Université de Picardie Jules Verne

Working Papers from HAL

Abstract: This paper delves into the intriguing world of Bitcoin price prediction techniques with a thorough introduction and analysis. Algorithms and quantitative models are now essential instruments for predicting Bitcoin price changes. These approaches use statistical and mathematical methods to examine historical price data and spot trends or patterns. Typical methods include machine learning models, which adjust and improve their predictions as they take in more data, and time series analysis, which forecasts future prices using past price data. The ARMA-GARCH models serve as the foundation for the conventional methods, which have been given preference in the relevant literature. These models are not perfect, however. As a result, several hybrid models combining the GARCH specification and neural network algorithm have evolved with the advent of machine learning algorithms. These models reduce error predictions, which is encouraging. In this work, we evaluate errors in the daily price forecast of Bitcoin for 2014-2024 using a univariate neural network GARCH model. Since the errors are smaller than those of a univariate traditional model for longer horizons, the results are consistent with the literature. They are helpful for investors, private financial actors, and relevant regulatory authorities.

Keywords: ARMA-GARCH process; Neural Network algorithms; Stock Market; Bitcoin; Volatility forecasts; MSC Classification Codes -: 62M45 62P05 (search for similar items in EconPapers)
Date: 2025-03-03
Note: View the original document on HAL open archive server: https://hal.science/hal-05241069v1
References: Add references at CitEc
Citations:

Downloads: (external link)
https://hal.science/hal-05241069v1/document (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:hal:wpaper:hal-05241069

Access Statistics for this paper

More papers in Working Papers from HAL
Bibliographic data for series maintained by CCSD ().