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

      Perbandingan Metode ARIMA dan Fuzzy Time Series untuk Peramalan Harga Saham Sektor Energi (Studi Kasus: PT XYZ)

      Thumbnail
      View/Open
      Fulltext (3.188Mb)
      Date
      2023
      Author
      Shodiqin, Tsamrat'z Zahra Alya Putri
      Masjkur, Mohammad
      Sulvianti, Itasia Dina
      Metadata
      Show full item record
      Abstract
      Peramalan merupakan hal penting untuk dapat memaksimalkan keuntungan dengan memprediksi di masa mendatang. Metode peramalan terus dikembangkan hingga dibagi menjadi metode konvensional dan soft-computing yang bertujuan untuk mendapatkan galat terkecil. Penelitian ini bertujuan untuk membandingkan kinerja prediksi metode ARIMA dan Fuzzy Time Series menggunakan model Saxena-Easo. Data yang digunakan adalah data harian dari PT XYZ sejak 1 Januari 2022 hingga 31 Desember 2022. Evaluasi kinerja model dilakukan dengan membandingkan nilai Mean Absolute Percentage Error (MAPE) untuk menentukan model yang lebih unggul. Hasilnya, baik model ARIMA(0,1,1) maupun Fuzzy Time Series merupakan model yang baik. MAPE yang dihasilkan oleh model ARIMA(0,1,1) sebesar 4,78% sementara Fuzzy Time Series dengan model Saxena-Easo sebesar 1,72%. Secara visual model ARIMA(0,1,1) menghasilkan peramalan garis lurus yang terus meningkat, sementara metode Fuzzy Time Series dengan model Saxena-Easo cenderung mengikuti pola data aktual. Sehingga didapatkan kesimpulan bahwa metode Fuzzy Time Series dengan model Saxena-Easo memiliki kinerja yang lebih baik daripada model ARIMA dalam meramal data non-linear.
       
      Forecasting is crucial for maximizing profits by predicting future prices. Forecasting methods have evolved, braching into conventional and soft-computing techniques aimed at minimizing prediction errors. The objective is to compare the predictive performance of ARIMA and Fuzzy Time Series models using the Saxena-Easo framework. Daily data from PT XYZ spanning from January 1, 2022, to December 31, 2022, is employed in this study. Model performance is assessed by comparing the Mean Absolute Percentage Error (MAPE) values to determine the superior model. Both the ARIMA(0,1,1) and Fuzzy Time Series models demonstrate strong predictive capabilities. ARIMA(0,1,1) yields a MAPE of 4.78%, while Fuzzy Time Series with the Saxena-Easo model achieves an even lower MAPE of 1.72%. Visually, ARIMA(0,1,1) generates a steadily increasing linear forecast, whereas the Fuzzy Time Series method with the Saxena-Easo model closely tracks the actual data pattern. Consequently, it can be concluded that the Fuzzy Time Series method with the Saxena-Easo model outperforms the ARIMA model in forecasting non-linear data.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/133437
      Collections
      • UT - Statistics and Data Sciences [2260]

      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