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.

      Gold Price Forecast through Bagging Exponential Smoothing Method

      Thumbnail
      View/Open
      Cover (331.5Kb)
      Full text (2.631Mb)
      Lampiran (749.6Kb)
      Date
      2022
      Author
      Atmaja, Arifah Afdila
      Fitrianto, Anwar
      Rahman, La Ode Abdul
      Metadata
      Show full item record
      Abstract
      Gold is one of the precious metals that remain popular among investors nowadays. It provides investors a safe haven when the economy is questionable. It is crucial to provide better information about the prospect of gold prices before investing. One way to do it in the statistical realm is through forecasting using the exponential smoothing method. In order to tackle challenges that may be encountered, an additional approach called bootstrap aggregating (bagging) was proposed. This study aims to compare the performance of exponential smoothing in a condition with and without bagging. Data that were used are gold prices from January 4th, 2010 until February 27th, 2022. As there is only sign of a trend component, it can be modelled using the Holt’s linear and the additive damped trend method. Models were evaluated based on MAPE and RMSE. Based on Wilcoxon signed-rank test, adding bagging is proven to bring statistically proven difference. For the daily and weekly series, the best model is Holt’s linear smoothing with bagging. While for the monthly series, the best one is Holt’s Linear smoothing without bagging. Lastly, it is predicted that gold prices will keep increasing until reaching more than Rp1000000/g in February 2023.
       
      Emas merupakan salah satu logam mulia yang popular di kalangan investor. Emas dapat menjadi alternatif investasi yang aman ketika kondisi perekonomian sedang lesu. Tentu peramalan harga emas ke depan menjadi poin pertimbangan bagi investor sebelum menanamkan modalnya. Dalam ranah statististik, salah satu metode yang dapat digunakan ialah pemulusan eksponensial. Untuk mengatasi tantangan yang timbul saat proses pemodelan, bootstrap aggregating diperkenalkan sebagai pendekatan tambahan. Penelitian ini bertujuan untuk membandingkan performa pemulusan eksponensial pada kondisi dengan dan tanpa bagging serta meramalkan harga emas satu tahun ke depan. Data yang digunakan ialah data harian harga emas dari 4 Januari 2010 s.d 27 Februari 2022. Tipe pemulusan eksponensial yang digunakan ialah Holt’s linear dan the additive damped trend method. Model kemudian dievaluasi berdasarkan MAPE dan RMSE. Hasil pengujian dengan Wilcoxon signed-rank test menunjukkan bahwa penggunaan bagging sebagai pendekaan tambahan terbukti memberikan hasil peramalan yang berbeda nyata. Untuk data harian dan mingguan, model terbaik ialah pemulusan Holt’s linear dengan bagging sedangkan untuk data bulanan, model terbaik ialah pemulusan Holt’s linear tanpa bagging. Pada bulan Februari 2023, diprediksi harga emas akan melampaui Rp1000000/g.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/113459
      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