Wisantoko, Bayuaji (2026) Peramalan Volatilitas Return Indeks LQ45 dengan Variabel Eksogen Return Nilai Tukar Rupiah Menggunakan Model ARIMAX-GARCHX dengan Intervensi Outlier pada Periode Pandemi COVID-19. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5003221024-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5003221024-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (4MB) | Request a copy |
Abstract
Indeks LQ45 sebagai representasi saham-saham berkapitalisasi besar dan likuid di Indonesia dipengaruhi oleh berbagai faktor makroekonomi, salah satunya nilai tukar Rupiah terhadap Dolar AS (USD/IDR). Hubungan tersebut menjadi semakin kompleks akibat pandemi COVID-19 yang memicu peningkatan volatilitas pasar, sehingga diperlukan model yang mampu memproyeksikan return dan volatilitas return secara simultan. Penelitian ini bertujuan untuk memodelkan dan meramalkan return serta volatilitas return Indeks LQ45 dengan mempertimbangkan pengaruh return nilai tukar Rupiah terhadap Dolar AS dan dampak guncangan pandemi COVID-19. Data yang digunakan berupa data harian periode 2 Januari 2019 hingga 2 Januari 2026. Analisis karakteristik data menunjukkan bahwa return Indeks LQ45 dan return nilai tukar Rupiah terhadap Dolar AS mengalami fenomena volatility clustering serta cenderung bergerak berlawanan arah. Pemodelan dilakukan menggunakan pendekatan ARIMAX-GARCHX dengan intervensi outlier melalui tahapan pendeteksian outlier pada periode pandemi COVID-19, pembentukan variabel dummy intervensi, prewhitening, dan identifikasi fungsi transfer. Model terbaik yang diperoleh adalah TF(0,0,0)-AR(2)-GARCH(1,1)-D_LQ45_t dengan distribusi Student-t. Hasil pemodelan menunjukkan bahwa perubahan return nilai tukar Rupiah terhadap Dolar AS berpengaruh langsung terhadap return Indeks LQ45 pada periode yang sama. Variabel dummy yang dibentuk dari 17 titik outlier signifikan selama pandemi COVID-19 mampu mengakomodasi pengaruh guncangan ekstrem. Evaluasi model dilakukan menggunakan pendekatan prediksi rekursif pada rentang data testing. Akurasi prediksi return diukur menggunakan Root Mean Square Error (RMSE), sedangkan akurasi prediksi volatilitas diukur menggunakan fungsi kerugian Quasi-Likelihood (QLIKE). Batas keseimbangan optimal model dicapai pada observasi ke-10 dengan RMSE 0,01302046 dan QLIKE -7,379757. Berdasarkan acuan tersebut, peramalan 10 hari ke depan menghasilkan point forecast return -0,000711 hingga 0,000078 dan proyeksi volatilitas 0,008218 hingga 0,009054. Model terbukti memadai memproyeksikan dinamika jangka pendek menuju keseimbangan jangka panjangnya.
=======================================================================================================================================
The LQ45 Index, representing Indonesia's large-cap and liquid stocks, is influenced by various macroeconomic factors, including the Rupiah to US Dollar (USD/IDR) exchange rate. This relationship became increasingly complex due to the COVID-19 pandemic, which triggered heightened market volatility, necessitating a model capable of simultaneously projecting returns and return volatility. This study aims to model and forecast LQ45 Index returns and return volatility, considering the USD/IDR exchange rate return's effect and COVID-19 pandemic shocks. Daily data from January 2, 2019, to January 2, 2026, are used. Data characteristic analysis shows that the LQ45 Index and Rupiah to US Dollar exchange rate returns experience volatility clustering and tend to move in opposite directions. Modeling uses the ARIMAX-GARCHX approach with outlier intervention through COVID-19 pandemic outlier detection, intervention dummy variable formation, prewhitening, and transfer function identification. The best model obtained is TF(0,0,0)-AR(2)-GARCH(1,1)-D_LQ45_t with a Student-t distribution. Modeling results indicate that changes in the Rupiah to US Dollar exchange rate return directly affect the LQ45 Index return within the same period. The dummy variable constructed from 17 significant outlier points during the COVID-19 pandemic adequately accommodates the impact of extreme shocks. Model evaluation uses a recursive prediction approach on the testing data range. Return prediction accuracy is measured using Root Mean Square Error (RMSE), while volatility prediction accuracy uses the Quasi-Likelihood (QLIKE) loss function. The model's optimal equilibrium limit is reached at the 10th observation with an RMSE of 0.01302046 and a QLIKE of -7.379757. Based on this reference, the 10-day ahead forecasting yields a point forecast return of -0,000711 to 0,000078 and a volatility projection of 0,008218 to 0,009054. The model proves adequate in projecting short-term dynamics toward its long-term equilibrium.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | ARIMAX-GARCHX, Indeks LQ45, Intervensi Outlier, Nilai Tukar Rupiah, Volatilitas Return, ARIMAX-GARCHX, LQ45 Index, Outlier Intervention, Return Volatility, Rupiah Exchange Rate |
| Subjects: | H Social Sciences > HA Statistics > HA30.3 Time-series analysis H Social Sciences > HB Economic Theory > Economic forecasting--Mathematical models. H Social Sciences > HG Finance > HG3881 Foreign exchange. H Social Sciences > HG Finance > HG4915 Stocks--Prices Q Science > QA Mathematics > QA280 Box-Jenkins forecasting R Medicine > RA Public aspects of medicine > RA644.C67 COVID-19 (Disease) |
| Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | BAYUAJI WISANTOKO |
| Date Deposited: | 30 Jul 2026 06:53 |
| Last Modified: | 30 Jul 2026 06:53 |
| URI: | http://repository.its.ac.id/id/eprint/139825 |
Actions (login required)
 |
View Item |