Aurellia, Patricia Juan and David Kaluge,, SE., MS., M.Ec.Dev., Ph.D (2023) Pemodelan Dan Peramalan Harga Cryptocurrency Dengan Model Markov Switching Autoregressive. Sarjana thesis, Universitas Brawijaya.
Abstract
Munculnya tren investasi cryptocurrency telah membawa jumlah pelanggan terdaftar aset kripto di Indonesia melebihi jumlah investor pasar modal. Meskipun peningkatan jumlah investor koin kripto terus terjadi, hal yang berbeda ditunjukkan oleh nilai transaksi yang terus menurun. Penurunan terjadi karena nilai koin kripto yang sangat fluktuatif berdampak pada keputusan investasi investor. Penelitian ini bertujuan guna mendapatkan model terbaik serta peramalan terkait harga cryptocurrency sehingga dapat meminimalisir rasa khawatir dan kerugian yang dapat dialami investor. Peneliti menggunakan harga penutupan 5 koin kripto dengan kapitalisasi terbesar. Penelitian ini menggunakan metode Markov Switching Autoregressive dengan alat analisis R Studio. Hasil penelitian menunjukkan model terbaik BTC adalah MS(3)AR(1), model BNB adalah MS(3)AR(1), model ETH adalah MS(3)AR(1), model XRP adalah MS(3)AR(2), dan untuk model ADA adalah MS(3)AR(2). Nilai RMSE menunjukkan bahwa BTC menjadi koin dengan akaurasi prediksi harga terbaik
English Abstract
The emergence of cryptocurrency investment trends has brought the number of registered customers for crypto assets in Indonesia to surpass the number of investors in the capital market. Despite the continuous increase in the number of cryptocurrency investors, a different scenario is depicted by the declining transaction values. This decrease is attributed to the highly volatile nature of cryptocurrency coins, which impacts investors' investment decisions. This research aims to obtain the best model and forecasts related to cryptocurrency prices in order to minimize concerns and potential losses experienced by investors. The researcher employed the closing prices of the five largest market-capitalized cryptocurrency coins. The research utilized the Markov Switching Autoregressive method with R Studio as the analytical tool. The research findings indicate that the best model for BTC is MS(3)AR(1), the model for BNB is MS(3)AR(1), the model for ETH is MS(3)AR(1), the model for XRP is MS(3)AR(2), and the model for ADA is MS(3)AR(2). The RMSE values indicate that BTC is the coin with the most accurate price prediction
| Item Type: | Thesis (Sarjana) |
|---|---|
| Identification Number: | :0523020456 |
| Subjects: | 300 Social sciences > 330 Economics |
| Divisions: | Fakultas Ekonomi dan Bisnis > Ilmu Ekonomi |
| Depositing User: | Unnamed user with username nova |
| Date Deposited: | 04 Jan 2024 02:41 |
| Last Modified: | 04 Jan 2024 02:41 |
| URI: | http://repository.ub.ac.id/id/eprint/205785 |
![[thumbnail of DALAM MASA EMBARRGO]](http://repository.ub.ac.id/style/images/fileicons/text.png) |
Text (DALAM MASA EMBARRGO)
Patricia Juan Aurellia.pdf Restricted to Registered users only until 31 December 2025. Download (3MB) |
Actions (login required)
 |
View Item |