Prediksi Harga Bitcoin Berdasarkan Harga Indeks S&P500 Menggunakan Metode Fungsi Transfer

Samudra, Ahmad Rizqi Perdana Putra (2024) Prediksi Harga Bitcoin Berdasarkan Harga Indeks S&P500 Menggunakan Metode Fungsi Transfer. Diploma thesis, Institut Teknologi Sepuluh Nopember.

Text 10611910000003-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2026. Download (1MB) | Request a copy

Abstract

Perkembangan indrustri teknologi terus berkembang dengan kemunculan teknologi dan inovasi yang baru sehingga banyak industri lain mendapatkan dampak positif salah satunya dalam bidang industri keuangan dan invetasi. Kemunculan teknologi blockchain membuat dunia keuangan dan investasi cukup mengalami perkembangan dimana orang dalam berinvestasi biasa memilih emas, saham dan sebagainya. Sejak kemunculan teknologi blockchain muncul aset baru berupa cryptocurrency dengan asset kripto pertama berupa bitcoin dengan capital gain yang cukup besar akan tetapi memliki risiko yang cukup besar karena fluktuasi yang cukup tinggi. Banyak faktor yang mempengaruhi peningkatan dan penurunan harga kripto, perkembangan perusahaan development kripto dan bahkan juga dipengaruhi makro ekonomi dunia. Sehingga perlu adanya model untuk memprediksi pergerakan harga Bitcoin terhadap instrument lain. Dalam hal ini harga indeks S&P500 cukup ber korelasi. Oleh karena itu penelitian ini akan memprediksi pergerakan harga bitcoin karena merupakan asset kripto yang mempunyai dominasi terbesar diramalkan dari harga index S&P500 yang merupakan index yang mewakili saham AS dimana Amerika merupakan movers teknologi makro ekonomi dunia, penelitian ini akan menggunakan fungsi transfer berdasarkan harga index S&P500 harga kemarin (closing price) untuk memprediksi harga bitcoin. Model fungsi transfer adalah model peramalan yang menggunakan variabel prediktor dan variabel respon berdasarkan pada runtun waktu, data yang digunakan adalah data sekunder yang diambil melalui website yahoo.finance, didapatkan hasil model untuk prediksi b=1, r=0, s=0 dengan deret noise ARIMA (1,0,0) menyatakan nilai ramalan untuk harga Bitcoin pada waktu ke-t dipengaruhi oleh harga 1 minggu sebelumnya, harga S&P500 pada 1 dan 2 minggu sebelumnya dengan nilai RMSE diketahui sebesar 1482 dan MAPE 7%. Hasil Prediksi harga Bitcoin dari 5 Juli – 20 Agustus 2023 menyatakan harga Bitcoin kedepannya akan mengalami sideways untuk beberapa bulan kedepan sepanjang tahun 2023.
================================================================================================================================
The development of the technology industry continues to grow with the emergence of new technology and innovation so that many other industries have a positive impact, one of which is in the financial and investment industry. The emergence of blockchain technology has made the world of finance and investment experience quite a development where people in investing usually choose gold, shares and so on. Since the emergence of blockchain technology, new assets have emerged in the form of cryptocurrency, with the first crypto asset being bitcoin with quite large capital gains, but it has quite large risks due to quite high fluctuations. Many factors influence the increase and decrease in crypto prices, the development of crypto development companies and are even influenced by world macroeconomics. So there is a need for a model to predict Bitcoin price movements against other instruments. In this case, the price of the S&P500 index is quite correlated. Therefore, this research will predict Bitcoin price movements because it is a crypto asset that has the greatest dominance, predicted from the price of the S&P500 index, which is an index that represents US shares, where America is the world's macroeconomic technology mover. This research will use a transfer function based on the S&P index price. 500 yesterday's price (closing price) to predict the price of bitcoin. The transfer function model is a forecasting model that uses predictor variables and response variables based on a time series. The data used is secondary data taken via the yahoo.finance website. The model results obtained for prediction are b=1, r=0, s=0 with a series ARIMA noise (1,0,0) states that the forecast value for the Bitcoin price at time t is influenced by the price 1 week previously, the price of the S&P500 in the previous 1 and 2 weeks with a known RMSE value of 1482 and MAPE 7%. The Bitcoin price prediction results from July 5 – August 20 2023 state that future Bitcoin prices will experience sideways for the next few months throughout 2023.

Item Type:	Thesis (Diploma)
Uncontrolled Keywords:	Bitcoin, S&P500, Transfer Function, cryptocurrency, fungsi transfer
Subjects:	H Social Sciences > HG Finance > HG4012 Mathematical models H Social Sciences > HG Finance > HG4910 Investments H Social Sciences > HG Finance > HG4915 Stocks--Prices
Divisions:	Faculty of Vocational > 49501-Business Statistics
Depositing User:	Ahmad Rizqi Perdana Putra Samudra
Date Deposited:	23 Feb 2024 08:23
Last Modified:	23 Feb 2024 08:24
URI:	http://repository.its.ac.id/id/eprint/107696

Actions (login required)

View Item