Pemodelan Vector Autoregressive – Dynamic Conditional Correlation Exponential Generalized Autoregressive Conditional Heteroscedasticity (VAR-DCC EGARCH)

Ranibaya, Yoni (2019) Pemodelan Vector Autoregressive – Dynamic Conditional Correlation Exponential Generalized Autoregressive Conditional Heteroscedasticity (VAR-DCC EGARCH). Sarjana thesis, Universitas Brawijaya.

Abstract

Model Vector Autoregressive (VAR) merupakan model yang terdiri dari beberapa variabel endogen yang memiliki hubungan dua arah dan bersifat stasioner. Salah satu asumsi pada model VAR(p) adalah ragam sisaan model yang bersifat konstan atau homogen. Apabila asumsi tersebut tidak terpenuhi akibat adanya volatilitas tinggi yang cenderung memiliki leverage effect (efek asimetris), maka perlu diantisipasi dengan menggunakan metode yang dapat memodelkan ragam sisaan. Salah satu metode yang dapat memodelkan ragam pada analisis data deret waktu multivariat yang memiliki efek asimetris adalah Dynamic Conditional Correlation Exponential Generalized Autoregressive Conditional Heteroscedasticity (DCC EGARCH). Penelitian ini bertujuan untuk memodelkan Indeks Harga Saham Gabungan (IHSG) dan Kurs Dolar Amerika menggunakan model VAR-DCC EGARCH(1,1). Hasil penelitian menunjukkan pada data IHSG dan Kurs Dolar Amerika terdapat hubungan dua arah, volatilitas asimetris dan memiliki korelasi yang berbeda di setiap waktu, sehingga didapatkan pemodelan VAR(1)-DCC EGARCH(1,1) bahwa IHSG dipengaruhi oleh Kurs Dolar Amerika satu bulan sebelumnya dan Kurs Dolar Amerika dipengaruhi oleh IHSG satu bulan sebelumnya.

English Abstract

Vector Autoregressive (VAR) is a model consists of several endogenous variables with a bidirectional relationship and stationary. One of the assumptions in the VAR(p) model is that there are constant or homogeneous variants of the model. If the assumption is violated due to high volatility which tends to have a leverage effect (asymmetric effect), it needs to be anticipated by using methods that can model the variance of the residuals. One method that can be used for variance modeling in multivariate time series data analysis that has asymmetrical effects is Dynamic Conditional Correlation Exponential Generalized Autoregressive Conditional Heteroscedasticity (DCC EGARCH). This study aims to model the Indonesia Composite Index and the US Dollar Exchange Rate using the VAR-DCC EGARCH model (1.1). The results showed in the Indonesia Composite Index data and the US Dollar Exchange Rate there is a two-way relationship, asymmetric volatility and has a different correlation at each time so that the VAR (1)-DCC EGARCH(1,1) modeling was obtained that the Indonesia Composite Index was influenced by one month US Dollar Exchange Rate previously and the US Dollar Exchange Rate was influenced by Indonesia Composite Index one month earlier.

Item Type: Thesis (Sarjana)
Identification Number: SKR/MIPA/2019/389/052001544
Uncontrolled Keywords: DCC EGARCH, IHSG, Kurs Dolar Amerika, VAR
Subjects: 500 Natural sciences and mathematics > 519 Probabilities and applied mathematics > 519.5 Statistical mathematics > 519.53 Descriptive statistics, multivariate analysis, analysis of variance and covariance > 519.536 Regression analysis
Divisions: Fakultas Matematika dan Ilmu Pengetahuan Alam > Statistika
Depositing User: Budi Wahyono Wahyono
Date Deposited: 10 Aug 2020 07:35
Last Modified: 10 Aug 2020 07:35
URI: http://repository.ub.ac.id/id/eprint/179948
Full text not available from this repository.

Actions (login required)

View Item View Item