IPB University Logo

SCIENTIFIC REPOSITORY

IPB University Scientific Repository collects, disseminates, and provides persistent and reliable access to the research and scholarship of faculty, staff, and students at IPB University

AI Repository
 
Building and Categories


      View Item 
      •   IPB Repository
      • Final Assignments
      • Undergraduate Final Assignments
      • UF - School of Data Science, Mathematic and Informatics
      • UF - Statistics and Data Sciences
      • View Item
      •   IPB Repository
      • Final Assignments
      • Undergraduate Final Assignments
      • UF - School of Data Science, Mathematic and Informatics
      • UF - Statistics and Data Sciences
      • View Item
      JavaScript is disabled for your browser. Some features of this site may not work without it.

      Penerapan Metode Gated Recurrent Unit (GRU) pada Peramalan Harga Cryptocurrency: Studi Kasus Bitcoin

      Thumbnail
      View/Open
      Cover (1.022Mb)
      Fulltext (1.040Mb)
      Lampiran (103.0Kb)
      Date
      2026
      Author
      HAMBALI, THARIQ
      Alamudi, Aam
      Silvianti, Pika
      Metadata
      Show full item record
      Abstract
      Bitcoin merupakan aset kripto dengan sistem terdesentralisasi yang memiliki karakteristik data deret waktu dinamis dan bervolatilitas cukup tinggi. Penelitian ini bertujuan mendapatkan konfigurasi hyperparameter terbaik pada arsitektur Gated Recurrent Unit (GRU) dan menerapkannya dalam peramalan harga harian Bitcoin selama 30 hari ke depan. Pengolahan komputasi memanfaatkan data harga Bitcoin periode 18 Juli 2010 hingga 30 September 2025 melalui pendekatan deret waktu univariate. Tahapan komputasi mencakup praproses Min-Max Scaling, transformasi sliding window, dan pencarian konfigurasi terbaik menggunakan algoritma Grid Search dengan mengombinasikan timesteps, neuron, batch size, dan epoch. Hasil komputasi menunjukkan arsitektur GRU paling optimal terbentuk pada kombinasi timesteps 180, 256 neuron, batch size 32, dan epoch 50. Konfigurasi tersebut memiliki kemampuan generalisasi yang baik tanpa indikasi overfitting, yang ditunjukkan oleh tingkat kesalahan prediksi Root Mean Squared Error (RMSE) sebesar 1.642,803 USD dan Mean Absolute Percentage Error (MAPE) sebesar 1,846%. Penerapan konfigurasi terbaik untuk peramalan 30 hari ke depan memproyeksikan kelanjutan tren penurunan harga secara bertahap pada rentang 114.000 - 103.000 USD. Pencapaian ini menyimpulkan bahwa model GRU univariate mampu memetakan arah kecenderungan harga secara umum, tetapi memiliki keterbatasan dalam melihat kejutan volatilitas harian pada data aktual di pasar.
       
      Bitcoin is a decentralized cryptocurrency asset characterized by dynamic and highly volatile time series data. This study aimed to obtain the best hyperparameter configuration for the Gated Recurrent Unit (GRU) architecture and apply it to forecast daily Bitcoin prices for the next 30 days. Computational processing utilized historical Bitcoin price data from July 18, 2010, to September 30, 2025, through a univariate time series approach. The computational stages included Min-Max Scaling preprocessing, sliding window transformation, and the search for the optimal configuration using the Grid Search algorithm by combining timesteps, neurons, batch size, and epochs. The computational results showed that the optimal GRU architecture was formed by a combination of 180 timesteps, 256 neurons, a batch size of 32, and 50 epochs. This configuration demonstrated good generalization capabilities without any indication of overfitting, as evidenced by a Root Mean Squared Error (RMSE) of 1,642.803 USD and a Mean Absolute Percentage Error (MAPE) of 1.846%. The application of the best configuration for the next 30 days forecasting projected a continuation of a gradual downward price trend within the range of 114,000 to 103,000 USD. These findings concluded that the univariate GRU model was capable of mapping the general direction of price trends but had limitations in capturing the daily volatility shocks of actual market data.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/173585
      Collections
      • UF - Statistics and Data Sciences [103]

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository
        

       

      Browse

      All of IPB RepositoryCollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

      My Account

      Login

      Application

      google store

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository