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

      Penanganan Keheterogenan Ragam Menggunakan Transformasi Box-Cox dan GARCH pada Model Deret Waktu ARIMA

      Thumbnail
      View/Open
      Cover (1.635Mb)
      Fullteks (3.063Mb)
      Lampiran (1.312Mb)
      Date
      2022
      Author
      Lestari, Lucky
      Budiarti, Retno
      Erliana, Windiani
      Metadata
      Show full item record
      Abstract
      Model Autoregressive Integrated Moving Average (ARIMA) merupakan sebuah metode peramalan yang menggunakan sifat dan karakteristik dari data masa lalu untuk meramalkan data deret waktu selanjutnya. Namun, diperlukan penyelesaian keheterogenan ragam pada data dengan melakukan transformasi Box-Cox atau Generalized Autoregressive Conditional Heteroskedasticity (GARCH) pada model ARIMA, sehingga diperoleh model ARIMA-GARCH. Selanjutnya dilakukan perbandingan antara model ARIMA yang melalui transfomasi Box-Cox dengan model ARIMA-GARCH dengan melihat besar error dari masing-masing model untuk kemudian dilakukan peramalan menggunakan salah satu model tersebut. Data yang digunakan adalah harga saham harian PT Telkom Indonesia Tbk tanggal 2 Januari 2018 hingga 28 Desember 2018, yang kemudian dilakukan peramalan untuk 6 periode mendatang. Model akhir yang diperoleh adalah ARIMA(2,1,2) untuk model dengan transformasi Box-Cox dengan MAPE sebesar 1.55%, yang mana lebih kecil dari model ARIMA(2,1,2)-GARCH(4,0) dengan MAPE sebesar 53.93%. Hasil peramalan untuk 6 periode berikutnya menghasilkan nilai MAPE sebesar 1.01%, di mana tingkat kesalahan yang dihasilkan cukup rendah.
       
      Autoregressive Integrated Moving Average model (ARIMA) is a forecasting method that uses characteristics of past data to predict future time series data. However, it is necessary to solve the heterogeneity of variance in the data by performing a Box-Cox transformation or Generalized Autoregressive Conditional Heteroskedasticity (GARCH) on the ARIMA model, so that the ARIMA-GARCH model is obtained. Furthermore, a comparison is made between the ARIMA model through the Box-Cox transformation and the ARIMA-GARCH model by looking at the error size of each model and then forecasting using one of these models. The data used is the daily share price of PT Telkom Indonesia Tbk from January 2, 2018 to December 28, 2018, which is then forecasted for the next 6 periods. The final model obtained is ARIMA(2,1,2) for the model with Box-Cox transformation with a MAPE of 1.55%, which is smaller than the ARIMA(2,1,2)-GARCH(4,0) model with a MAPE of 53.93%. Forecasting results for the next 6 periods produce a MAPE value of 1.01%, in which the error rate is quite low.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/110571
      Collections
      • UT - Actuaria [205]

      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