Analisis Dependensi Return Antar-Saham Sektor Perbankan Menggunakan Metode Mixed Copula

Siahaan, Keisha Alessandra Lynn (2025) Analisis Dependensi Return Antar-Saham Sektor Perbankan Menggunakan Metode Mixed Copula. Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 5006211087-Undergraduate_Thesis.pdf] Text
5006211087-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only

Download (4MB) | Request a copy

Abstract

Investasi merupakan aktivitas penempatan dana dengan tujuan memperoleh imbal hasil di masa depan. Salah satu instrumen investasi yang banyak diminati oleh investor adalah saham, karena menawarkan potensi capital gain serta dividen. Namun, instrumen ini juga memiliki tingkat risiko yang tinggi akibat volatilitas pasar yang tidak dapat diprediksi secara pasti. Dalam menghadapi risiko inheren tersebut, diversifikasi portofolio menjadi strategi penting yang perlu diterapkan. Diversifikasi yang efektif tidak hanya dilakukan dengan menambah jumlah saham dalam portofolio, tetapi juga dengan mempertimbangkan korelasi antar saham, di mana idealnya dipilih saham-saham dengan korelasi rendah atau negatif untuk menurunkan risiko total portofolio secara signifikan dibandingkan risiko individual masing-masing saham. Penelitian ini bertujuan untuk meminimalkan risiko investasi melalui pendekatan Mixed Copula, yang dipilih karena kemampuannya dalam memodelkan ketergantungan antar saham secara fleksibel, tidak terbatas pada hubungan linier, dan dapat menangkap karakteristik pergerakan ekstrem yang kerap terjadi di pasar keuangan. Data yang digunakan dalam penelitian ini berupa harga saham harian dari empat bank besar di Indonesia—BBCA, BBNI, BBRI, dan BMRI—dengan periode observasi dari 1 Januari 2019 hingga 31 Desember 2024. Hasil estimasi menunjukkan bahwa untuk seluruh pasangan saham, model Mixed Copula Clayton-Gumbel merupakan model terbaik yang mampu menggambarkan struktur ketergantungan secara akurat. Pada seluruh pasangan saham yang dianalisis, model ini memberikan hasil konsisten di mana bobot Copula Gumbel cenderung sedikit lebih dominan dibandingkan Copula Clayton. Hal ini mengindikasikan bahwa ketergantungan antar saham secara umum lebih sensitif terhadap pergerakan ekstrem ke atas dibandingkan ke bawah. Setelah model copula terbaik diperoleh, pengukuran ketergantungan dilakukan menggunakan koefisien Kendall’s Tau. Hasilnya menunjukkan nilai antara 0,307 hingga 0,426 untuk seluruh pasangan saham. Nilai tersebut mencerminkan adanya hubungan yang kuat, positif, dan searah antar saham-saham yang dianalisis, di mana pergerakan harga pada satu saham cenderung diikuti oleh pergerakan harga yang sama arah pada saham lainnya. Penelitian ini menyajikan wawasan penting bagi para investor dalam upaya meminimalkan risiko melalui strategi diversifikasi portofolio, dengan menekankan pertimbangan cermat atas karakteristik hubungan antar saham dan pemanfaatan model copula yang paling sesuai dengan karakteristik data spesifik.
===================================================================================================================================
Investment is the activity of placing funds with the aim of obtaining returns in the future. One of the most popular investment instruments for investors is stocks, as they offer the potential for capital gains and dividends. However, this instrument also carries a high level of risk due to market volatility that cannot be predicted with certainty. In the face of such inherent risks, portfolio diversification is an important strategy that needs to be implemented. Effective diversification is not only done by increasing the number of stocks in the portfolio, but also by considering the correlation between stocks, where ideally stocks with low or negative correlation are selected to significantly reduce the total risk of the portfolio compared to the individual risk of each stock. This study aims to minimize investment risk through the Mixed Copula approach, which was chosen for its ability to model dependence between stocks flexibly, not limited to linear relationships, and can capture the characteristics of extreme movements that often occur in financial markets. The data used in this study are daily stock prices of four major banks in Indonesia-BBCA, BBNI, BBRI, and BMRI-with the observation period from January 1, 2019 to December 31, 2024. The estimation results show that for all pairs of stocks, the Clayton-Gumbel Mixed Copula model is the best model capable of accurately describing the dependency structure. For all pairs of stocks analyzed, the model provides consistent results where the weight of the Gumbel Copula tends to be slightly more dominant than the Clayton Copula. This indicates that the dependency between stocks is generally more sensitive to extreme upward movements than downward ones. Once the best copula model is obtained, a dependency measurement is performed using Kendall's Tau coefficient. The results show values between 0.307 and 0.426 for all pairs of stocks. These values reflect a strong, positive and unidirectional relationship between the stocks analyzed, where price movements in one stock tend to be followed by price movements in the same direction in other stocks. This research provides important insights for investors in their efforts to minimize risk through portfolio diversification strategies, emphasizing careful consideration of the characteristics of the relationship between stocks and the utilization of copula models that best fit the specific data characteristics.

Item Type: Thesis (Other)
Uncontrolled Keywords: portfolio diversification, banking stocks, Mixed Copula, Kendall’s Tau, stock dependence, diversifikasi portofolio, saham perbankan, Mixed Copula, Kendall’s Tau, ketergantungan antar saham
Subjects: Q Science > QA Mathematics > QA276 Mathematical statistics. Time-series analysis. Failure time data analysis. Survival analysis (Biometry)
Q Science > QA Mathematics > QA401 Mathematical models.
Divisions: Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis
Depositing User: Keisha Alessandra Lynn Siahaan
Date Deposited: 28 Jul 2025 02:40
Last Modified: 28 Jul 2025 02:40
URI: http://repository.its.ac.id/id/eprint/121906

Actions (login required)

View Item View Item