Analisis prediktif volatilitas indeks harga saham gabungan menggunakan garch dan lstm dengan pendekatan volume-up strategy = Predictive analysis of Indonesia composite index volatility using garch and lstm with the volume-up strategy approach

Dharmaningrat, I Made Adhi (2024) Analisis prediktif volatilitas indeks harga saham gabungan menggunakan garch dan lstm dengan pendekatan volume-up strategy = Predictive analysis of Indonesia composite index volatility using garch and lstm with the volume-up strategy approach. Bachelor thesis, Universitas Pelita Harapan.

	Text (Title) Title.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (176kB)
	Text (Abstract) Abstract.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (804kB)
	Text (ToC) ToC.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (789kB)
	Text (Chapter1) Chapter1.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (1MB)
	Text (Chapter2) Chapter2.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (4MB)
	Text (Chapter3) Chapter3.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (1MB)
	Text (Chapter4) Chapter4.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (2MB)
	Text (Chapter5) Chapter5.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (693kB)
	Text (Bibliography) Bibliography.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (702kB)
	Text (Appendices) Appendices.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (33MB)

Abstract

Keterhubungan ekonomi antarnegara menjadi fokus utama dalam memahami dampak pergerakan pasar modal global terhadap indeks saham domestik. Dalam penelitian terdahulu, pergerakan nilai indeks saham global seperti Dow Jones, S&P 500, Hang Seng, dan Nikkei 225 terbukti memiliki dampak kausalitas terhadap pergerakan nilai IHSG di Indonesia. Tujuan penelitian ini adalah untuk menganalisis bagaimana realized volatility IHSG dapat diprediksi dengan menggunakan gabungan model GARCH dan LSTM, dengan memanfaatkan pendekatan Volume-Up Strategy. Melalui proses pengembangan model dengan melakukan optimisasi hyperparameter, kemudian evaluasi dilakukan untuk membandingkan performa berbagai model dan menganalisis dampak penggunaan VU Strategy dengan berbagai variasi nilai α, yaitu {10 1 , 10 2 ,..., 9 10 }, terhadap model GARCH-LSTM. Selain itu, proses optimisasi hyperparameter menghasilkan kombinasi optimal pada fungsi aktivasi, jumlah epoch, dan ukuran batch size. Penelitian ini menunjukkan bahwa model GARCH-LSTM memberikan prediksi volatilitas yang lebih akurat daripada model LSTM tunggal. Terutama, penerapan VU Strategy secara signifikan meningkatkan performa model GARCH-LSTM, dengan nilai α = 0,5 memberikan hasil terbaik dengan tingkat galat yang paling rendah. Hal ini menunjukkan bahwa integrasi antara model GARCH-LSTM dan VU Strategy memiliki potensi besar dalam meningkatkan akurasi prediksi volatilitas pasar keuangan, yang dapat mendukung pengambilan keputusan dalam berinvestasi. / The interconnection of economies between countries is the main focus in understanding the impact of global capital market movements on domestic stock indices. Previous research have shown a direct correlation between fluctuations in global stock indices such as Dow Jones, S&P 500, Hang Seng, and Nikkei 225, and the movement of JKSE value in Indonesia. The aim of this study is to analyze how the realized volatility of JKSE can be predicted using a combination of GARCH and LSTM models, utilizing the Volume-Up Strategy approach. Through the process of model development and hyperparameter optimization, evaluations are conducted to compare the performance of various models and analyze the impact of using the VU Strategy with various values of α, which are {10 1 , 10 2 ,..., 9 10 }, on the GARCH-LSTM model. Furthermore, the hyperparameter optimization process yields an optimal combination of activation function, epoch, and batch size. The study demonstrates that the GARCH-LSTM model provides more accurate volatility predictions compared to the single LSTM model. Particularly, the implementation of the VU Strategy significantly improves the performance of the GARCH-LSTM model, with α = 0.5 yielding the best results with the lowest error rate. This indicates that the integration of the GARCH-LSTM model and the VU Strategy has great potential in enhancing the accuracy of financial market volatility predictions, which can support decision-making in investment.

Actions (login required)

View Item