Peramalan Nilai Ekspor Migas di Provinsi Kalimantan Timur Menggunakan Metode Berbasis Long Short Term Memory

Putra, Jehezkiel Pratamavions Permata (2025) Peramalan Nilai Ekspor Migas di Provinsi Kalimantan Timur Menggunakan Metode Berbasis Long Short Term Memory. Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 5003211155-Undergraduate_Thesis.pdf] Text
5003211155-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only

Download (7MB) | Request a copy

Abstract

Indonesia adalah salah satu negara dengan kekayaan sumber daya alam melimpah, termasuk minyak dan gas bumi (migas). Provinsi Kalimantan Timur merupakan salah satu penyumbang terbesar ekspor migas di Indonesia, meskipun tren nilai ekspornya menurun dalam beberapa tahun terakhir. Penelitian ini bertujuan untuk meramalkan nilai ekspor migas di Kalimantan Timur menggunakan metode LSTM, dan model hybrid ARIMA-LSTM. Data yang digunakan adalah data bulanan nilai ekspor migas dari Januari 2011 hingga Desember 2024, dengan evaluasi model berdasarkan metrik RMSE dan MAPE. Berdasarkan hasil evaluasi, model LSTM memberikan hasil terbaik dengan nilai RMSE sebesar 65,44. Penelitian ini diharapkan dapat memberikan referensi strategis bagi pemerintah daerah dalam merencanakan kebijakan terkait stabilisasi dan peningkatan nilai ekspor migas di Kalimantan Timur.
====================================================================================================================================
Indonesia is one of the countries rich in natural resources, including oil and gas. East Kalimantan Province is one of the largest contributors to oil and gas exports in Indonesia, although the export value has shown a declining trend in recent years. This study aims to forecast the oil and gas export values in East Kalimantan using the LSTM method and the hybrid ARIMA-LSTM model. The data used consists of monthly oil and gas export values from January 2011 to December 2024, with model evaluation based on RMSE and MAPE metrics. Based on the evaluation results, the LSTM model produced the best performance with an RMSE value of 65.44. This research is expected to provide strategic references for local governments in planning policies related to the stabilization and enhancement of oil and gas export values in East Kalimantan.

Item Type: Thesis (Other)
Uncontrolled Keywords: ARIMA, LSTM, hybrid model, peramalan, ekspor migas, Kalimantan Timur, ARIMA, LSTM, hybrid model, forecasting, oil and gas exports, East Kalimantan
Subjects: Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines.
Divisions: Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User: Jehezkiel Pratamavions Permata Putra
Date Deposited: 01 Aug 2025 10:00
Last Modified: 01 Aug 2025 10:00
URI: http://repository.its.ac.id/id/eprint/126269

Actions (login required)

View Item View Item