A Machine Learning Approach for Bitcoin Forecasting
Bitcoin is one of the cryptocurrencies that has gained popularity in recent years. Previous studies have shown that closing price alone is not enough to forecast its future level, and other price-related features are necessary to improve forecast accuracy. We introduce a new set of time series and d...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-06-01
|
| Series: | Engineering Proceedings |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2673-4591/39/1/27 |
| _version_ | 1827726280954478592 |
|---|---|
| author | Stefano Sossi-Rojas Gissel Velarde Damian Zieba |
| author_facet | Stefano Sossi-Rojas Gissel Velarde Damian Zieba |
| author_sort | Stefano Sossi-Rojas |
| collection | DOAJ |
| description | Bitcoin is one of the cryptocurrencies that has gained popularity in recent years. Previous studies have shown that closing price alone is not enough to forecast its future level, and other price-related features are necessary to improve forecast accuracy. We introduce a new set of time series and demonstrate that a subset is necessary to improve directional accuracy based on a machine learning ensemble. In our experiments, we study which time series and machine learning algorithms deliver the best results. We found that the most relevant time series that contribute to improving directional accuracy are open, high, and low, with the largest contribution of low in combination with an ensemble of a gated recurrent unit network and a baseline forecast. The relevance of other Bitcoin-related features that are not price-related is negligible. The proposed method delivers similar performance to the state of the art when observing directional accuracy. |
| first_indexed | 2024-03-10T22:47:53Z |
| format | Article |
| id | doaj.art-0231e6d77fcb4e74a29604d665928390 |
| institution | Directory Open Access Journal |
| issn | 2673-4591 |
| language | English |
| last_indexed | 2024-03-10T22:47:53Z |
| publishDate | 2023-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Engineering Proceedings |
| spelling | doaj.art-0231e6d77fcb4e74a29604d6659283902023-11-19T10:30:41ZengMDPI AGEngineering Proceedings2673-45912023-06-013912710.3390/engproc2023039027A Machine Learning Approach for Bitcoin ForecastingStefano Sossi-Rojas0Gissel Velarde1Damian Zieba2Computational Systems Engineering, Universidad Privada Boliviana, Cochabamba 3967, BoliviaComputational Systems Engineering, Universidad Privada Boliviana, Cochabamba 3967, BoliviaFaculty of Economic Sciences, University of Warsaw, 00927 Warsaw, PolandBitcoin is one of the cryptocurrencies that has gained popularity in recent years. Previous studies have shown that closing price alone is not enough to forecast its future level, and other price-related features are necessary to improve forecast accuracy. We introduce a new set of time series and demonstrate that a subset is necessary to improve directional accuracy based on a machine learning ensemble. In our experiments, we study which time series and machine learning algorithms deliver the best results. We found that the most relevant time series that contribute to improving directional accuracy are open, high, and low, with the largest contribution of low in combination with an ensemble of a gated recurrent unit network and a baseline forecast. The relevance of other Bitcoin-related features that are not price-related is negligible. The proposed method delivers similar performance to the state of the art when observing directional accuracy.https://www.mdpi.com/2673-4591/39/1/27Bitcoinforecastingtime seriesmachine learning |
| spellingShingle | Stefano Sossi-Rojas Gissel Velarde Damian Zieba A Machine Learning Approach for Bitcoin Forecasting Engineering Proceedings Bitcoin forecasting time series machine learning |
| title | A Machine Learning Approach for Bitcoin Forecasting |
| title_full | A Machine Learning Approach for Bitcoin Forecasting |
| title_fullStr | A Machine Learning Approach for Bitcoin Forecasting |
| title_full_unstemmed | A Machine Learning Approach for Bitcoin Forecasting |
| title_short | A Machine Learning Approach for Bitcoin Forecasting |
| title_sort | machine learning approach for bitcoin forecasting |
| topic | Bitcoin forecasting time series machine learning |
| url | https://www.mdpi.com/2673-4591/39/1/27 |
| work_keys_str_mv | AT stefanosossirojas amachinelearningapproachforbitcoinforecasting AT gisselvelarde amachinelearningapproachforbitcoinforecasting AT damianzieba amachinelearningapproachforbitcoinforecasting AT stefanosossirojas machinelearningapproachforbitcoinforecasting AT gisselvelarde machinelearningapproachforbitcoinforecasting AT damianzieba machinelearningapproachforbitcoinforecasting |