View Item 
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - Faculty of Mathematics and Natural Sciences
      • UT - Physics
      • View Item
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - Faculty of Mathematics and Natural Sciences
      • UT - Physics
      • View Item
      JavaScript is disabled for your browser. Some features of this site may not work without it.

      Prediksi Dinamika Harga Penutupan Bitcoin Menggunakan Random Forest dengan Pendekatan Hukum Hooke

      Thumbnail
      View/Open
      Cover (605.5Kb)
      Fulltext (1.847Mb)
      Lampiran (832.4Kb)
      Date
      2025
      Author
      Hottua, Randy Ardiles
      Hardhienata, Hendradi
      Husin, Abd. Djamil
      Metadata
      Show full item record
      Abstract
      Bitcoin merupakan aset digital dengan volatilitas tinggi yang membuat prediksi harga menjadi tantangan. Penelitian ini bertujuan membangun model prediksi harga penutupan Bitcoin menggunakan algoritma random forest serta mengevaluasi pengaruh penambahan fitur konstanta Bitcoin berdasarkan transformasi Hukum Hooke. Data yang digunakan adalah harga harian Bitcoin dari tahun 2016 hingga 2024. Metode penelitian mencakup pra-pemrosesan data, ekstraksi indikator teknikal, penambahan fitur konstanta, tuning parameter dengan GridSearchCV, dan evaluasi model menggunakan MAPE dan AUC. Hasil menunjukkan bahwa penambahan fitur konstanta meningkatkan akurasi prediksi harga, dengan nilai MAPE uji menurun dari 7,96% menjadi 7,74%. Namun, pada model klasifikasi arah pergerakan, penambahan fitur tidak memberikan peningkatan performa dan justru menurunkan nilai AUC dari 0,94 menjadi 0,89. Temuan ini menunjukkan bahwa fitur konstanta Bitcoin lebih efektif digunakan dalam model regresi dibandingkan klasifikasi.
       
      Bitcoin is a digital asset with high volatility, making price prediction a challenge. This study aims to develop a Bitcoin closing price prediction model using the random forest algorithm and evaluate the effect of adding a Bitcoin constant feature derived from a transformation of Hooke’s Law. The dataset consists of daily Bitcoin prices from 2016 to 2024. The research includes data preprocessing, technical indicator extraction, constant feature construction, parameter tuning using GridSearchCV, and model evaluation using MAPE and AUC. The results show that adding the constant feature improves price prediction accuracy, as the test MAPE decreased from 7,96% to 7,74%. However, for price movement direction classification, the feature did not improve performance and instead lowered the AUC from 0,94 to 0,89. These findings indicate that the Bitcoin constant feature is more effective for regression models than for classification tasks.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/171042
      Collections
      • UT - Physics [1227]

      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