Peramalan dan dekomposisi untuk mata uang kripto dengan model facebook prophet
Cryptocurrencies are becoming one of the hottest topics in Indonesia's society. One of those issues concerns investors who incur financial losses as a result of investing in crypto. The facebook Prophet model, one of the forecast models, can offer a solution to this problem. The Prophet model i...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Department of Mathematics FMIPA University of Jember
2024-03-01
|
| Series: | Majalah Ilmiah Matematika dan Statistika |
| Online Access: | https://jurnal.unej.ac.id/index.php/MIMS/article/view/39159 |
| _version_ | 1827300554677682176 |
|---|---|
| author | Dany Rahman Dewi Rachmatin Rini Marwati |
| author_facet | Dany Rahman Dewi Rachmatin Rini Marwati |
| author_sort | Dany Rahman |
| collection | DOAJ |
| description | Cryptocurrencies are becoming one of the hottest topics in Indonesia's society. One of those issues concerns investors who incur financial losses as a result of investing in crypto. The facebook Prophet model, one of the forecast models, can offer a solution to this problem. The Prophet model is built using four function. These variables are trend, seasonality, holidays, and additional regressions. The Prophet model benefits from a number of advantages, one of which is its ability to generate decomposition graphs. The decomposition may give analysts more insight into the data they are analyzing. The Prophet model is used to forecast and decompose the price of a cryptocurrenciy called Solana in this study. A multiplicative model with linear function as trend function, weekly seasonality, and daily seasonality as seasonality function is the best model for Solana price forecasting and decomposition. Additionally, hyperparameters in the model are tuned so the model won’t suffer underfitting or overfitting indications. The fitted Prophet model is good at forecasting as a result of the evaluation process. As a result of the forecast and decomposition, the forecasted value and the decomposition graph of the Solana exchange rate for one hour later show that the price of Solana will remain constant.
Keywords: Cryptocurrency, time series, forecasting, decomposition, facebook prophet
MSC2020: 62M10 |
| first_indexed | 2024-04-24T16:06:54Z |
| format | Article |
| id | doaj.art-ccb5988d726c4b99a987cdc2c2682121 |
| institution | Directory Open Access Journal |
| issn | 1411-6669 2722-9866 |
| language | English |
| last_indexed | 2024-04-24T16:06:54Z |
| publishDate | 2024-03-01 |
| publisher | Department of Mathematics FMIPA University of Jember |
| record_format | Article |
| series | Majalah Ilmiah Matematika dan Statistika |
| spelling | doaj.art-ccb5988d726c4b99a987cdc2c26821212024-04-01T03:57:46ZengDepartment of Mathematics FMIPA University of JemberMajalah Ilmiah Matematika dan Statistika1411-66692722-98662024-03-01241486010.19184/mims.v24i1.3915939159Peramalan dan dekomposisi untuk mata uang kripto dengan model facebook prophetDany Rahman0Dewi Rachmatin1Rini Marwati2Departemen Pendidikan Matematika, FPMIPA, Universitas Pendidikan IndonesiaDepartemen Pendidikan Matematika, FPMIPA, Universitas Pendidikan IndonesiaDepartemen Pendidikan Matematika, FPMIPA, Universitas Pendidikan IndonesiaCryptocurrencies are becoming one of the hottest topics in Indonesia's society. One of those issues concerns investors who incur financial losses as a result of investing in crypto. The facebook Prophet model, one of the forecast models, can offer a solution to this problem. The Prophet model is built using four function. These variables are trend, seasonality, holidays, and additional regressions. The Prophet model benefits from a number of advantages, one of which is its ability to generate decomposition graphs. The decomposition may give analysts more insight into the data they are analyzing. The Prophet model is used to forecast and decompose the price of a cryptocurrenciy called Solana in this study. A multiplicative model with linear function as trend function, weekly seasonality, and daily seasonality as seasonality function is the best model for Solana price forecasting and decomposition. Additionally, hyperparameters in the model are tuned so the model won’t suffer underfitting or overfitting indications. The fitted Prophet model is good at forecasting as a result of the evaluation process. As a result of the forecast and decomposition, the forecasted value and the decomposition graph of the Solana exchange rate for one hour later show that the price of Solana will remain constant. Keywords: Cryptocurrency, time series, forecasting, decomposition, facebook prophet MSC2020: 62M10https://jurnal.unej.ac.id/index.php/MIMS/article/view/39159 |
| spellingShingle | Dany Rahman Dewi Rachmatin Rini Marwati Peramalan dan dekomposisi untuk mata uang kripto dengan model facebook prophet Majalah Ilmiah Matematika dan Statistika |
| title | Peramalan dan dekomposisi untuk mata uang kripto dengan model facebook prophet |
| title_full | Peramalan dan dekomposisi untuk mata uang kripto dengan model facebook prophet |
| title_fullStr | Peramalan dan dekomposisi untuk mata uang kripto dengan model facebook prophet |
| title_full_unstemmed | Peramalan dan dekomposisi untuk mata uang kripto dengan model facebook prophet |
| title_short | Peramalan dan dekomposisi untuk mata uang kripto dengan model facebook prophet |
| title_sort | peramalan dan dekomposisi untuk mata uang kripto dengan model facebook prophet |
| url | https://jurnal.unej.ac.id/index.php/MIMS/article/view/39159 |
| work_keys_str_mv | AT danyrahman peramalandandekomposisiuntukmatauangkriptodenganmodelfacebookprophet AT dewirachmatin peramalandandekomposisiuntukmatauangkriptodenganmodelfacebookprophet AT rinimarwati peramalandandekomposisiuntukmatauangkriptodenganmodelfacebookprophet |