PERAMALAN HARGA SAHAM MENGGUNAKAN HYBRID ARFIMA DAN GJR-GARCH: Studi Kasus Harga Saham Tesla, Inc

    Yemima Imanuela Setyadi, - and Fitriani Agustina, - and Nar Herrhyanto, - (2025) PERAMALAN HARGA SAHAM MENGGUNAKAN HYBRID ARFIMA DAN GJR-GARCH: Studi Kasus Harga Saham Tesla, Inc. S1 thesis, Universitas Pendidikan Indonesia.

    Abstract

    Pasar saham merupakan salah satu instrumen investasi yang diminati, dengan saham Tesla, Inc (TSLA) sebagai salah satu saham yang populer di kalangan investor. Investor dapat memperoleh keuntungan dalam bentuk capital gain. Fluktuasi harga saham mendorong investor untuk melakukan analisis pergerakan saham sebagai dasar pengambilan keputusan investasi. Namun, analisis tersebut tidak mudah dilakukan sehingga peramalan harga saham merupakan aspek yang penting dalam investasi saham. Dengan demikian, penelitian ini bertujuan untuk memperoleh model peramalan terbaik dan hasil peramalan harga saham Tesla, Inc dengan menggunakan hybrid ARFIMA dan GJR-GARCH yang menangani ketergantungan jangka panjang pada harga saham dan volatilitas asimetris. Penggabungan kedua model tersebut diharapkan dapat meningkatkan akurasi peramalan harga saham. Data yang digunakan merupakan data harga penutupan harian saham Tesla, Inc dengan periode Juli 2022 - Februari 2025, yang diolah menggunakan Python dan R. Berdasarkan hasil pengolahan data, diperoleh model terbaik yaitu ARFIMA(1,0.4995373,1)-GJR-GARCH(2,1)-Skewed Student-t yang memperoleh nilai akurasi peramalan MAPE sebesar 4.543% pada data latih, 6.073% pada data uji, dan 1.745% pada hasil peramalan lima periode ke depan. Selain itu, dilakukan juga perhitungan Value at Risk dengan tujuan memperoleh hasil perhitungan risiko. Pada tingkat kepercayaan 90%, 95%, dan 99%, diperoleh kerugian terburuk tidak akan lebih dari 4.19%, 5.75%, dan 9.65% dari total investasi. The stock market is one of the most popular investment instruments, with Tesla, Inc. (TSLA) being one of the most favored stocks among investors. Investors can earn profits in the form of capital gains. Stock price fluctuations encourage investors to analyze stock movements as a basis for making investment decisions. However, such analysis is not easy to perform, making stock price forecasting an important aspect of stock investment. Therefore, this study aims to obtain the best forecasting model and the forecasting results of Tesla, Inc stock prices using a hybrid ARFIMA and GJR-GARCH to handle long-term dependence in stock prices and asymmetric volatility. The combination of these two models is expected to improve forecasting accuracy. The data used in this study are daily closing prices of Tesla, Inc. stock for the period from July 2022 to February 2025, processed using Python and R. Based on the data analysis, the best model obtained is ARFIMA(1,0.4995373,1)-GJR-GARCH(2,1)-Skewed Student-t, which achieves forecasting accuracy with MAPE values of 4.543% for the training data, 6.073% for the testing data, and 1.745% for the five-step-ahead forecast. In addition, a Value at Risk calculation was conducted to obtain the estimated risk. At confidence levels of 90%, 95%, and 99%, the worst potential losses are estimated not to exceed 4.19%, 5.75%, and 9.65% of the total investment.

    Baca Full Text klik disini
    [thumbnail of S_MAT_2102276_Title.pdf] Text
    S_MAT_2102276_Title.pdf
    Download (673kB)
    [thumbnail of S_MAT_2102276_Chapter1.pdf] Text
    S_MAT_2102276_Chapter1.pdf
    Download (447kB)
    [thumbnail of S_MAT_2102276_Chapter2.pdf] Text
    S_MAT_2102276_Chapter2.pdf
    Restricted to Staf Perpustakaan
    Download (578kB) | Request a copy
    [thumbnail of S_MAT_2102276_Chapter3.pdf] Text
    S_MAT_2102276_Chapter3.pdf
    Download (815kB)
    [thumbnail of S_MAT_2102276_Chapter4.pdf] Text
    S_MAT_2102276_Chapter4.pdf
    Restricted to Staf Perpustakaan
    Download (1MB) | Request a copy
    [thumbnail of S_MAT_2102276_Chapter5.pdf] Text
    S_MAT_2102276_Chapter5.pdf
    Download (419kB)
    [thumbnail of S_MAT_2102276_Appendix.pdf] Text
    S_MAT_2102276_Appendix.pdf
    Restricted to Staf Perpustakaan
    Download (1MB) | Request a copy
    Official URL: https://repository.upi.edu/
    Item Type: Thesis (S1)
    Additional Information: ID SInta Dosen pembimbing: Fitriani Agustina: 5981275
    Uncontrolled Keywords: Peramalan, Saham TSLA, ARFIMA, GJR-GARCH Forecasting, TSLA Stock, ARFIMA, GJR-GARCH
    Subjects: H Social Sciences > H Social Sciences (General)
    H Social Sciences > HF Commerce
    L Education > L Education (General)
    Divisions: Fakultas Pendidikan Matematika dan Ilmu Pengetahuan Alam > Program Studi Matematika - S1 > Program Studi Matematika (non kependidikan)
    Depositing User: Yemima Imanuela Setyadi
    Date Deposited: 30 Jul 2025 06:36
    Last Modified: 30 Jul 2025 06:36
    URI: http://repository.upi.edu/id/eprint/134899

    Actions (login required)

    View Item View Item