AN INVESTIGATION OF INDONESIA’S TRANSITION TOWARDS RENEWABLE ENERGY USING TIME SERIES ANALYSIS

The purpose of this study is to showcase modern forecasting methodologies for predicting the path of renewable energy-based electricity generation for Indonesia to attain the U.N.’s Sustainable Development Goals (SDG) by 2030. For this forecasting technique, the study uses 40 years of annual previou...

Full description

Saved in:
Bibliographic Details
Published inThe Journal of energy and development Vol. 47; no. 1/2; pp. 297 - 326
Main Authors Manni, Umme Humayara, Datuk. Dr. Kasim Hj. Md. Mansur
Format Journal Article
LanguageEnglish
Published Boulder International Research Center for Energy and Economic Development (ICEED) 01.04.2022
International Research Center for Energy and Economic Development
Subjects
Carbon dioxide
Data
Economic growth
Electricity generation
Emissions
Energy consumption
Forecasting
Fossil fuels
Natural gas
Policy making
Production
Renewable energy
Sustainable development
Time series
Indonesia
Online AccessGet full text

Cover

More Information
Summary:The purpose of this study is to showcase modern forecasting methodologies for predicting the path of renewable energy-based electricity generation for Indonesia to attain the U.N.’s Sustainable Development Goals (SDG) by 2030. For this forecasting technique, the study uses 40 years of annual previous data from secondary data sources. It is critical for policy makers to understand how to apply historical data with time series and machine learning approaches to different industries in Indonesia. In this study, robust Holt-Winter’s forecasting techniques and decision tree regressions were used. The results show that forecasting of renewable energy generation in Indonesia is generally moderate, and that more renewable energy-based power production is required to meet SDG goal 7, which is to ensure access to affordable, reliable, sustainable, and modern energy for all. The findings of this study can be used to anticipate renewable energy capacity and generation, in particular, as well as other factors such as CO2 emissions, fossil fuel reserves, and other energy-related variables, in general. The main contributions of this research are twofold: first, it introduces the process of applying conventional and machine learning techniques for projections of time series data, and, second, where most previous studies in this type of research use descriptive discussion, this study utilizes an empirical approach to pinpoint the accuracy of prediction techniques. The absence of daily data on renewable energy-based electricity production in Indonesia, which could be used to increase the accuracy of these models, is a drawback of this study. Other ASEAN countries can use the same forecasting technique for renewable energy generation.
ISSN:0361-4476
2831-9168