Kusumadewi, Nabila Sinta (2026) Peramalan Nilai Ekspor Bahan Bakar Mineral Indonesia Ke Asia Timur Berdasarkan Pengaruh Pandemi Dan Konflik Geopolitik Menggunakan Model Hybrid Time Series Regression Generalized Space-Time Autoregressive (TSR-GSTAR). Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5003221100-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5003221100-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (9MB) | Request a copy |
Abstract
Komoditas bahan bakar mineral yang tergolong dalam Harmonized System (HS) 27 merupakan salah satu komoditas ekspor utama Indonesia dengan kontribusi sekitar 20% terhadap total nilai ekspor nasional pada tahun 2024. China, Jepang, dan Korea Selatan merupakan negara tujuan utama ekspor komoditas tersebut sehingga dinamika nilai ekspornya berpotensi saling memengaruhi. Selain itu, pandemi COVID-19 dan konflik Rusia–Ukraina diduga memengaruhi dinamika nilai ekspor melalui perubahan pasar energi global. Penelitian ini bertujuan mendeskripsikan karakteristik nilai ekspor bahan bakar mineral Indonesia ke kawasan Asia Timur, memperoleh model hybrid Time Series Regression–Generalized Space-Time Autoregressive (TSR-GSTAR) terbaik, serta menghasilkan peramalan nilai ekspor selama 12 bulan ke depan. Data yang digunakan berupa data bulanan nilai ekspor bahan bakar mineral Indonesia ke China, Jepang, dan Korea Selatan periode Januari 2014 hingga Maret 2026. Analisis dilakukan menggunakan statistika deskriptif, regresi time series dengan variabel tren, dummy pandemi COVID-19, dan dummy konflik Rusia–Ukraina, kemudian dilanjutkan dengan pemodelan GSTAR pada residual regresi menggunakan bobot seragam, invers jarak, dan normalisasi korelasi silang. Model terbaik dipilih berdasarkan nilai RMSE terkecil dengan mempertimbangkan pengujian asumsi residual. Hasil penelitian menunjukkan bahwa ketiga negara tujuan memiliki pola pergerakan nilai ekspor yang berbeda. Variabel tren dan dummy konflik Rusia–Ukraina berpengaruh signifikan terhadap nilai ekspor pada ketiga negara tujuan, sedankan variabel dummy pandemi COVID-19 hanya berpengaruh signifikan pada China dan Jepang. Model terbaik yang diperoleh adalah hybrid TSR-GSTAR(1_1) restricted dengan bobot invers jarak karena menghasilkan nilai RMSE out-sample dan rerata RMSE terkecil serta memenuhi asumsi residual white noise dan normalitas. Hasil peramalan menunjukkan bahwa nilai ekspor ke China diperkirakan meningkat secara bertahap, sedangkan nilai ekspor ke Jepang dan Korea Selatan diperkirakan mengalami penurunan pada awal periode peramalan sebelum kembali meningkat. Secara keseluruhan, China diproyeksikan tetap menjadi negara tujuan utama ekspor bahan bakar mineral Indonesia selama periode peramalan.
========================================================================================================================================
Mineral fuel commodities classified under Harmonized System (HS) Code 27 are among Indonesia's major export commodities, contributing approximately 20% of the country's total export value in 2024. China, Japan, and South Korea are the primary export destinations, with export dynamics that are potentially interconnected. The export dynamics of these countries are potentially have been affected by major global events, particularly the COVID-19 pandemic and the Russia–Ukraine conflict. This study aims to develop the best hybrid Time Series Regression–Generalized Space-Time Autoregressive (TSR-GSTAR) model for forecasting Indonesia's mineral fuel export values to East Asia. Monthly export value data from January 2014 to March 2026 were analyzed. The proposed hybrid approach combines time series regression, incorporating trend and intervention variables representing the COVID-19 pandemic and the Russia–Ukraine conflict, with GSTAR modeling of the regression residuals using uniform, inverse distance, and normalized cross-correlation (NCC) spatial weights. Model performance was evaluated based on the Root Mean Square Error (RMSE) and residual diagnostic tests. The results indicate that the trend and the Russia–Ukraine conflict significantly affect export values in all destination countries, while the COVID-19 pandemic significantly affects exports to China and Japan. Among the evaluated models, the restricted hybrid TSR-GSTAR〖(1〗_1) with inverse distance weights provides the best forecasting performance, yielding the lowest out-sample and average RMSE while satisfying the white noise and normality assumptions. Forecasts for the next 12 months suggest a gradual increase in exports to China, whereas exports to Japan and South Korea are expected to decline initially before recovering. These findings demonstrate that the proposed hybrid TSR-GSTAR model effectively captures both temporal dynamics and spatial dependence in Indonesia's mineral fuel exports, with China projected to remain the largest export destination in East Asia.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Bahan Bakar Mineral, Hybrid TSR-GSTAR, Nilai Ekspor, Peramalan, Spasial-Temporal, Export Value, Forecasting, Mineral Fuel, Spatio-Temporal |
| Subjects: | H Social Sciences > HA Statistics H Social Sciences > HB Economic Theory |
| Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | Nabila Sinta Kusumadewi |
| Date Deposited: | 31 Jul 2026 07:52 |
| Last Modified: | 31 Jul 2026 07:52 |
| URI: | http://repository.its.ac.id/id/eprint/139686 |
Actions (login required)
 |
View Item |