Pemodelan Markov Switching Garch (Penerapan Pada Return Index Dowjones)

Hadiyat, Mochammad Arbi (2007) Pemodelan Markov Switching Garch (Penerapan Pada Return Index Dowjones). Masters thesis, Institut Teknologi Sepuluh November.

Text 1305201004-Master_Thesis.pdf Restricted to Repository staff only Download (13MB)

Abstract

Sejak dikemukakan oleh Engle (1982) dan Boilers lev (1986), model ARCH-GARCH telah banyak digunakan untuk mendeskripsikan perilalcu volatilitas suatu time series, terutama pada data-data tentang saham dan indeks berjangka. Salah satu kekurangan dalam model ARCH-GARCH adalah ketidakmarnpuannya untuk melihat transisi atau perubahan perilaku antara volatilitas rendah dengan volatilitas tinggi. Dalam penelitian ini, markov switching GARCH dikaji dan diterapkan untuk melihat adanya regime volatilitas yang berbeda, yakni regime volatilitas rendah dan regime volatilitas tinggi pada data return indeks Dowjones. Namun, tidak ada informasi yang menunjukkan bahwa suatu observasi dimiliki oleh regime pertama atau regime kedua, dan untuk mengatasinya digunakan algoritma EM untuk estimasi parameter. Hasil yang diperoleh menunjukkan bahwa return indeks Dowjones memang mempunyai dua regime volatilitas, hal ini ditandai dengan signifikansi parameter GARCH pada masing-masing regime. Melalui matrik transisi, regime volatilitas rendah temyata lebih banyak terjadi pada return indeks Dowjones dibandingkan regime volatilitas tinggi.
====================================================================================================================================
Since firstly proposed by Engle (1982) and Bollerslev (1986), ARCH-GARCH models have been used to describe volatility behaviors of time series, especially in stock market analysis. One of the weaknesses of ARCH-GARCH is its inability to model behavior transition between high volatilities and low volatilities. In this research, markov switching GARCH is investigated and applied to capture the presence of different volatility regimes, i.e. low volatilities regime and high volatility regime in Dowjones index return. However, there is no infonnation to decide which observations belong to each of the regimes, and to account this difficulty, EM algorithm is applied for parameter estimation. The result shows that Dowjones index return includes two volatility regimes. The transition matrix of the model yields that low volatility regime is often happened than the high one.

Item Type:	Thesis (Masters)
Additional Information:	RTSt 519.233 Had p
Uncontrolled Keywords:	Algoritma EM, GARCH, Markov Chain, Markov Switching GARCH, Volatilitas, EM algorithm, GARCH, Markov Chain, Markov Switching GARCH, Volatility
Subjects:	Q Science > QC Physics > QC174.17.M33 Markov processes
Divisions:	Faculty of Mathematics and Science > Statistics > 49101-(S2) Master Thesis
Depositing User:	magang .
Date Deposited:	01 Oct 2026 02:58
Last Modified:	01 Oct 2026 02:58
URI:	http://repository.its.ac.id/id/eprint/145122

Actions (login required)

View Item