Peramalan Harga Cryptocurrency Menggunakan Vector Autoregressive Model

Aries, Abdillah Nasikh (2024) Peramalan Harga Cryptocurrency Menggunakan Vector Autoregressive Model. Diploma thesis, Institut Teknologi Sepuluh Nopember.

Text 10611910000058_Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (1MB) | Request a copy

Abstract

Bidang ekonomi terus mengalami inovasi selama uang masih digunakan alat pembayaran. Salah satu perkembangan teknologi dalam bidang ekonomi adalah cryptocurrency. Saat ini aset kripto menjadi instrument investasi yang berhasil menarik minat masyarakat Indonesia karena mampu menjanjikan potensi imbal balik dengan hasil yang tinggi, namun risiko yang dihadapi juga meningkat dikarenakan pergerakan harga di pasar cenderung bergerak dengan fluktuasi yang tinggi. Penelitian ini bertujuan untuk rekomendasi cryptocurrency kepada para investor dengan cara membandingkan beberapa coin berdasarkan marketcap. Penelitian ini menggunakan model VAR karena hasil peramalan lebih baik dari peramalan dengan metode persamaan simultan yang kompleks. Vectorautoregresive mengembangkan model secara bersamaan di dalam suatu sitem yang kompleks multivariate, sehingga dapat menangkap hubungan keseluruhan variabel dalam persamaan tersebut. Hasil penelitian ini adalah data belum stationer sehingga perlu differencing satu kali agar data stasioner. Model terbaik dari data stationer adalah VAR (2,1) dengan tingkat kesalahan RMSE sebesar 22.018,25. Model untuk meramal harga cryptocurrency pada tanggal 1 Juli 2023 hingga 31 Juli 2023 adalah VAR (2,1) menghasilkan pergerakan yang cenderung konstan.
========================================================================================================================
The field of economics continues to experience innovation as long as money is still used as a means of payment. One technological development in the economic field is cryptocurrency. Currently, crypto assets have become an investment instrument that successfully attracts the interest of the Indonesian public because they promise high potential returns, but the associated risks also increase due to the high fluctuations in market prices. This research aims to provide cryptocurrency recommendations to investors by comparing several coins based on market cap. The study employs the VAR (Vector Autoregressive) model because the forecasting results are better than those obtained from simultaneous equation methods that are complex. Vector Autoregressive develops models simultaneously within a complex multivariate system, enabling it to capture the overall relationships among the variables in the equation. The research findings indicate that the data is not stationary, requiring differencing once to make it stationary. The best model for stationary data is VAR (2,1) with a Root Mean Square Error (RMSE) of 22,018.25. The model for forecasting cryptocurrency prices from July 1, 2023, to July 31, 2023, is VAR (2,1), which produces relatively constant movements.

Item Type:	Thesis (Diploma)
Uncontrolled Keywords:	Cryptocurrency, Marketcap, Multivariat, Vector autoregressive
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines.
Divisions:	Faculty of Vocational > 49501-Business Statistics
Depositing User:	Aries abdillah nasikh
Date Deposited:	12 Aug 2024 04:06
Last Modified:	12 Aug 2024 04:06
URI:	http://repository.its.ac.id/id/eprint/114785

Actions (login required)

View Item