IPB University Logo

SCIENTIFIC REPOSITORY

IPB University Scientific Repository collects, disseminates, and provides persistent and reliable access to the research and scholarship of faculty, staff, and students at IPB University

AI Repository
 
Building and Categories


      View Item 
      •   IPB Repository
      • Final Assignments
      • Undergraduate Final Assignments
      • UF - School of Data Science, Mathematic and Informatics
      • UF - Statistics and Data Sciences
      • View Item
      •   IPB Repository
      • Final Assignments
      • Undergraduate Final Assignments
      • UF - School of Data Science, Mathematic and Informatics
      • UF - Statistics and Data Sciences
      • View Item
      JavaScript is disabled for your browser. Some features of this site may not work without it.

      I-Optimal Saturated Design dengan Modified Fedorov's Exchange Algorithm

      Thumbnail
      View/Open
      Cover (2.457Mb)
      Fulltext (1.027Mb)
      Lampiran (263.0Kb)
      Date
      2026
      Author
      Yusuf, Rakesha Putra Antique
      Oktarina, Sachnaz Desta
      Syafitri, Utami Dyah
      Metadata
      Show full item record
      Abstract
      Tingginya biaya operasional eksperimen menuntut efisiensi melalui saturated design, di mana jumlah satuan percobaan sama dengan parameter model. Terdapat dua tujuan dalam regresi, yaitu pendugaan parameter dan pendugaan ragam prediksi. Penelitian ini bertujuan untuk mencari rancangan yang optimal dari aspek pendugaan ragam prediksi namun dengan jumlah runs yang sama dengan jumlah parameter model dengan I-Optimality Criterion. Untuk mendapatkan rancangan tersebut digunakan Modified Fedorov’s Exchange Algorithm yang merupakan salah satu dari point exchange algorithm yang membutuhkan candidate set. Penelitian ini akan digunakan tiga candidate set, yaitu Central Composite Design (CCD), titik titik dalam daerah percobaan dari rancangan faktorial, dan gabungan, dievaluasi berdasarkan nilai I-Optimality Criterion dan I-Efficiency. Untuk menghindari singularitas matriks dalam perhitungan I-Optimality Criterion ditambahkan regularisasi ??. I-Optimal Saturated Design untuk dua faktor dengan candidate set gabungan diperoleh 6 titik rancangan dengan I-Optimality Criterion terendah sebesar 0,7361. Dibandingkan dengan candidate set CCD, maka rancangan tersebut mempunyai I-Efficiency sebesar 106,75%. Pada rancangan IOptimal Design untuk 3 faktor diperoleh juga candidate set gabungan memiliki IOptimality Criterion terendah sebesar 1,3866 dengan 10 titik rancangan dan IEfficiency mencapai 115,54% terhadap rancangan CCD.
       
      High operational costs in experiments demand efficiency through a saturated design, where the number of experimental runs equals the model parameters. Regression has two main objectives: parameter estimation and prediction variance estimation. This study aims to find an optimal design focusing on the prediction variance estimation, with a number of runs exactly equal to the number of model parameters, using the I-Optimality Criterion. To obtain this design, the Modified Fedorov’s Exchange Algorithm was used, which is a point exchange algorithm that requires a candidate set. This research utilizes three candidate sets—Central Composite Design (CCD), factorial, and a combined set—which are evaluated based on their I-Optimality Criterion and I-Efficiency values. To avoid matrix singularity in calculating the I-Optimality Criterion, a regularization ?? is added. The I-Optimal Design for two factors using the combined candidate set yielded 6 design points with the lowest I-Optimality Criterion of 0.7361. Compared to the CCD candidate set, this design achieved the I-Efficiency of 106.76%. Similarly, for the 3-factor I-Optimal Design, the combined candidate set also achieved the lowest I-Optimality Criterion of 1.3866 with 10 design points, and an I-Efficiency reaching 115.54% relative to the CCD design.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/173683
      Collections
      • UF - Statistics and Data Sciences [103]

      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