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

      Penerapan Regresi Nonparametrik Spline Truncated dan B-Spline Dalam Memodelkan Faktor-Faktor Yang Memengaruhi Indeks Harga Saham Gabungan

      Thumbnail
      View/Open
      Cover (603.7Kb)
      Fulltext (947.4Kb)
      Lampiran (605.9Kb)
      Date
      2025
      Author
      Hamid, Yogi Nur
      Silvianti, Pika
      Firdawanti, Aulia Rizki
      Metadata
      Show full item record
      Abstract
      Regresi bekerja sangat baik pada data yang telah diketahui polanya, namun tidak semua data dapat diketahui polanya sehingga perlu adanya pendekatan fungsi tertentu. Dalam kasus ini, pendekatan nonparametrik digunakan sebagai alternatif dari regresi parametrik. Salah satu fungsi yang sering digunakan pada pendekatan regresi nonparameterik yaitu spline. Fungsi spline memiliki fleksibilitas yang sangat baik dalam menduga kurva regresi dengan pola yang berubah-ubah. Kasus ini sering kali dijumpai pada data perekonomian, salah satunya IHSG. IHSG mencerminkan kondisi kinerja saham di pasar modal dan sangat bergantung pada faktor internal dan eksternal. Penelitian ini bertujuan menerapkan pendekatan regresi nonparametrik spline untuk memodelkan IHSG berdasarkan faktor-faktor yang memengaruhinya. Basis fungsi yang digunakan yaitu truncated power dan B-spline. Titik knot optimal ditentukan berdasarkan kriteria Generalized Cross Validation (GCV) minimum. Pendugaan parameter ditentukan menggunakan metode Ordinary Least Square (OLS) dengan orde linier. Hasil menunjukkan bahwa model spline truncated dan B-spline mampu menangani permasalahan terkait fleksibilitas dalam menangkap pola hubungan nonlinier dengan nilai adjusted ??2 masing-masing 85,32% dan 91,14%. Model terbaik yaitu model B-spline dengan nilai MAPE latih dan uji terendah masing-masing sebesar 3,01% dan 5,63%. Berdasarkan uji signifikansi parameter, diperoleh empat peubah yang berpengaruh signifikan terhadap IHSG, yaitu inflasi, DJIA, harga emas dunia, dan harga minyak dunia.
       
      Regression works effectively when the data pattern is known. However, not all data exhibit clear patterns, thus requiring specific functional approaches. In such cases, nonparametric regression is employed as an alternative to parametric regression. One of the most commonly used functions in nonparametric regression is the spline function, which offers excellent flexibility in estimating regression curves with varying patterns. This condition frequently arises in economic data, including the Jakarta Composite Index (JCI), which reflects stock market performance and is highly influenced by both internal and external factors. This study aims to apply a nonparametric spline regression approach to model the JCI based on its influencing factors. The basis functions used are truncated power spline and B-spline. The optimal knot positions are determined using the minimum Generalized Cross Validation (GCV) criterion. Parameter estimation is performed using the Ordinary Least Squares (OLS) method with linear order. The results show that both truncated spline and B-spline models are capable of addressing flexibility in capturing nonlinear relationships, with adjusted ??2 values of 85.32% and 91.14%, respectively. The best performing model is the B-spline model, achieving the lowest training and testing MAPE values of 3.01% and 5.63%, respectively. Based on the parameter significance test, four variables are found to significantly influence the JCI: inflation, DJIA, world gold price, and world oil price.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/165119
      Collections
      • UT - Statistics and Data Sciences [83]

      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