Scholarlink

<p>Global electricity consumption has grown steadily, driven by increasing household, business, and industry demand. In Indonesia, the rapid growth in electricity consumption has led to significant challenges, including an oversupply of electricity in recent years. To address this issue and create more efficient policies, obtaining accurate data on household electricity consumption at more granular levels is crucial. However, the data collected by Statistics Indonesia (BPS) through the National Socio-Economic Survey (SUSENAS) is limited to reliable district or city-level estimates due to sample size constraints. Therefore, alternative methods and data sources are needed to provide accurate estimates at the sub-district level. This study employs the Small Area Estimation (SAE) method, utilizing additional data from significant data sources and Village Potential Statistics (PODES) to estimate household electricity consumption per capita at the sub-district level. The results demonstrate that the Empirical Best Linear Unbiased Prediction (EBLUP), Spatial EBLUP, and Hierarchical Bayes (HB) Lognormal SAE models offer more accurate and precise estimates than traditional direct estimation methods. Among these, the HB Lognormal SAE model was identified as the most effective based on the evaluation of the results.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Application of small area estimation using big data sources to estimate electricity consumption per capita households (case study: sub-district level in Central Java, Indonesia)

  • 期刊论文

  • 发表日期：

    2025年9月8日

查看全文
  • Raflizal Fikrar Odriansyah,
  • Setia Pramana,
  • Amiek Chamami

摘要

Global electricity consumption has grown steadily, driven by increasing household, business, and industry demand. In Indonesia, the rapid growth in electricity consumption has led to significant challenges, including an oversupply of electricity in recent years. To address this issue and create more efficient policies, obtaining accurate data on household electricity consumption at more granular levels is crucial. However, the data collected by Statistics Indonesia (BPS) through the National Socio-Economic Survey (SUSENAS) is limited to reliable district or city-level estimates due to sample size constraints. Therefore, alternative methods and data sources are needed to provide accurate estimates at the sub-district level. This study employs the Small Area Estimation (SAE) method, utilizing additional data from significant data sources and Village Potential Statistics (PODES) to estimate household electricity consumption per capita at the sub-district level. The results demonstrate that the Empirical Best Linear Unbiased Prediction (EBLUP), Spatial EBLUP, and Hierarchical Bayes (HB) Lognormal SAE models offer more accurate and precise estimates than traditional direct estimation methods. Among these, the HB Lognormal SAE model was identified as the most effective based on the evaluation of the results.