IMPLEMENTASI PEMILIHAN PORTOFOLIO OPTIMAL MENGGUNAKAN METODE KLASTERISASI K-MEANS DAN MEAN VARIANCE EFFICIENT PORTOFOLIO (MVEP): studi kasus indeks saham LQ45

    Evonia R. Hutajulu, - and Dewi Rachmatin, - and Fitriani Agustina, - (2025) IMPLEMENTASI PEMILIHAN PORTOFOLIO OPTIMAL MENGGUNAKAN METODE KLASTERISASI K-MEANS DAN MEAN VARIANCE EFFICIENT PORTOFOLIO (MVEP): studi kasus indeks saham LQ45. S1 thesis, Universitas Pendidikan Indonesia.

    Abstract

    Investasi saham merupakan penanaman modal di pasar modal dengan tujuan memperoleh keuntungan, namun disertai risiko yang perlu dikelola. Salah satu strategi untuk meminimalkan risiko adalah membentuk portofolio, yaitu kombinasi saham yang dipilih secara strategis guna memaksimalkan imbal hasil pada tingkat risiko tertentu. Dalam kondisi pasar yang fluktuatif, diversifikasi risiko dalam pemilihan portofolio optimal menjadi sangat penting. Penelitian ini menerapkan metode klasterisasi hibrida dan Mean Variance Efficient Portofolio (MVEP) untuk membentuk dan mengoptimalkan portofolio yang efisien. Sebanyak 36 saham dari Indeks LQ45 periode Januari 2023 hingga Desember 2024 dikelompokkan berdasarkan karakteristik historis imbal hasil dan risiko menggunakan algoritma Centroid Linkage dan K-Means Clustering. Selanjutnya, dibentuk kombinasi portofolio beda klaster, sama klaster, dan tanpa klasterisasi, yang kemudian dioptimalkan menggunakan MVEP dengan kendala larangan short-selling (w_i>0) dan total bobot portofolio sama dengan satu. Evaluasi kinerja menggunakan sharpe ratio menunjukkan bahwa portofolio dari klaster yang sama memiliki kinerja terbaik dengan nilai sharpe ratio tertinggi sebesar 0,9034. Sementara itu, portofolio dari klaster sama menunjukkan efektivitas diversifikasi lebih baik dengan rata-rata korelasi antar saham terendah sebesar 0,0024. Portofolio tanpa klasterisasi menunjukkan performa terendah dengan nilai sharpe ratio sebesar −0,5312. Temuan ini mengindikasikan bahwa penerapan klasterisasi meningkatkan efektivitas optimisasi portofolio menggunakan MVEP, serta memperkuat pentingnya diversifikasi berdasarkan karakteristik saham. Stock investment is a form of capital placement in the capital market with the aim of generating future returns, but it also involves risks that must be managed. One strategy to minimize risk is by constructing a portfolio, which is a strategic combination of selected stocks aimed at maximizing returns at a certain level of risk. In a volatile market environment, risk diversification in the selection of an optimal portfolio becomes essential. This study applies a hybrid clustering method and the Mean Variance Efficient Portfolio (MVEP) approach to construct and optimize an efficient portfolio. A total of 36 stocks from the LQ45 Index for the period January 2023 to December 2024 were grouped based on historical return and risk characteristics using the Centroid Linkage and K-Means Clustering algorithms. Subsequently, portfolio combinations were formed across clusters, within clusters, and without clustering, and then optimized using MVEP under the constraints of no short-selling (w_i>0) and the total portfolio weight equal to one. Performance evaluation using the Sharpe Ratio shows that the portfolio formed within the same cluster achieved the best performance, with the highest Sharpe Ratio of 0,9034. Meanwhile, the cross-cluster portfolio demonstrated better diversification effectiveness with the lowest average inter-stock correlation of 0,0024. The non-clustered portfolio showed the weakest performance with a Sharpe Ratio of −0,5312. These findings indicate that the application of clustering enhances the effectiveness of portfolio optimization using MVEP and reinforces the importance of diversification based on stock characteristics.

    Baca Full Text klik disini
    [thumbnail of S_MAT_2109414_Title.pdf] Text
    S_MAT_2109414_Title.pdf
    Download (1MB)
    [thumbnail of S_MAT_2109414_Chapter1.pdf] Text
    S_MAT_2109414_Chapter1.pdf
    Download (309kB)
    [thumbnail of S_MAT_2109414_Chapter2.pdf] Text
    S_MAT_2109414_Chapter2.pdf
    Restricted to Staf Perpustakaan
    Download (554kB)
    [thumbnail of S_MAT_2109414_Chapter3.pdf] Text
    S_MAT_2109414_Chapter3.pdf
    Download (557kB)
    [thumbnail of S_MAT_2109414_Chapter4.pdf] Text
    S_MAT_2109414_Chapter4.pdf
    Restricted to Staf Perpustakaan
    Download (956kB)
    [thumbnail of S_MAT_2109414_Chapter5.pdf] Text
    S_MAT_2109414_Chapter5.pdf
    Download (349kB)
    [thumbnail of S_MAT_2109414_Appendix.pdf] Text
    S_MAT_2109414_Appendix.pdf
    Restricted to Staf Perpustakaan
    Download (882kB)
    Official URL: https://repository.upi.edu/
    Item Type: Thesis (S1)
    Additional Information: https://scholar.google.com/citations?user=BPoT-D4AAAAJ&hl=en ID SINTA Dosen Pembimbing: Dewi Rachmatin: 5975775 Fitriani Agustina: 5981275
    Uncontrolled Keywords: Optimisasi Portofolio, K-Means, Centroid Linkage, MVEP Portfolio Optimization, K-Means, Centroid Linkage, MVEP
    Subjects: H Social Sciences > HG Finance
    H Social Sciences > HJ Public Finance
    L Education > L Education (General)
    Divisions: Fakultas Pendidikan Matematika dan Ilmu Pengetahuan Alam > Program Studi Matematika - S1 > Program Studi Matematika (non kependidikan)
    Depositing User: Evonia R. Hutajulu
    Date Deposited: 31 Jul 2025 07:34
    Last Modified: 31 Jul 2025 07:34
    URI: http://repository.upi.edu/id/eprint/134967

    Actions (login required)

    View Item View Item