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

      Perbandingan Kinerja Model ARIMA dan k-Nearest Neighbor dalam Peramalan Harga Beras Indonesia

      Thumbnail
      View/Open
      Cover (698.3Kb)
      Fulltext (1.059Mb)
      Lampiran (366.2Kb)
      Date
      2025
      Author
      TAZKIA, AVINA
      Bukhari, Fahren
      Mayyani, Hidayatul
      Metadata
      Show full item record
      Abstract
      Harga beras di Indonesia merupakan indikator penting stabilitas ekonomi dan kesejahteraan masyarakat, sehingga fluktuasinya dapat menyebabkan ketidakpastian. Penelitian ini membandingkan kinerja model Autoregressive Integrated Moving Average (ARIMA) dan k-Nearest Neighbor (KNN) dalam meramalkan harga beras di Indonesia menggunakan data bulanan periode Januari 2010 hingga Maret 2024. Data dibagi menjadi 80% training dan 20% testing. Model ARIMA(1,1,0) dipilih sebagai model terbaik berdasarkan uji formal dan kriteria AIC, sementara kombinasi parameter KNN terbaik adalah k = 8, algoritma ‘auto’, metrik Manhattan, dan pembobotan ‘uniform’. Hasil menunjukkan ARIMA memiliki akurasi lebih baik dengan nilai MAPE dan RMSE lebih rendah, serta lebih efektif dalam menangkap tren jangka panjang meskipun kurang responsif terhadap fluktuasi mendadak. KNN lebih fleksibel dalam mengenali pola data, namun kurang stabil dalam menghadapi volatilitas tinggi. Oleh karena itu, ARIMA lebih cocok untuk data linear dan stabil, sedangkan KNN lebih sesuai untuk data yang lebih kompleks.
       
      Indonesian rice prices are an essential indicator of economic stability and public welfare, so their fluctuations can cause uncertainty. This study compares the performance of the Autoregressive Integrated Moving Average (ARIMA) and k-Nearest Neighbor (KNN) models in forecasting rice prices in Indonesia using monthly data from January 2010 to March 2024. The data is divided into 80% training and 20% testing. The ARIMA(1,1,0) model was selected as the best model based on the formal test and AIC criteria, while the best KNN parameter combination was k = 8, the 'auto' algorithm, Manhattan metric, and 'uniform' weighting. This study reveals that ARIMA outperforms KNN, yielding lower MAPE and RMSE values, and effectively capturing long-term trends. However, ARIMA is less responsive to sudden fluctuations. Conversely, KNN excels in recognizing complex patterns but struggles with high volatility. Thus, ARIMA suits linear and stable data, while KNN is better suited for complex data.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/160665
      Collections
      • UT - Mathematics [1487]

      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