Jurnal Matematika Sains dan Teknologi
Vol. 26 No. 2 (2025)

Hierarchical Bayes Application for Small Area Estimation with Error Measurement on Child Poverty in Sumatera Island

Aisha Arthamevia (Politeknik Statistika STIS)
Zahra Rizky Fadilah (Politeknik Statistika STIS)
Rasya Az Zahra (Politeknik Statistika STIS)
Sausan Salsabila (Politeknik Statistika STIS)
Renandika Salsabiila Agistasari (Politeknik Statistika STIS)
Nofita Istiana (Politeknik Statistika STIS)



Article Info

Publish Date
30 Sep 2025

Abstract

Child poverty on Sumatera Island remains a significant issue, as four provinces recorded child poverty rates above the national average in 2021, increasing to five provinces in 2022. To support more effective and targeted policies, reliable estimates at the district/city level are required; however, direct estimates from the March 2023 Susenas data showed low precision, with 37 of 154 districts/cities having a Relative Standard Error (RSE) greater than 25%. To improve accuracy, this study applied Small Area Estimation using a Hierarchical Bayes model with Measurement Error on the Beta distribution (SAE HB ME Beta). Empirical findings revealed serious precision problems, particularly in Kepulauan Bangka Belitung, where all districts had RSE values above 25%, and in West Sumatera, which ranked second with more districts exceeding the threshold than those below it, including the highest overall RSE. When the model was initially estimated jointly for all provinces, one district in West Sumatera still had an RSE above 25% and estimates for Kepulauan Bangka Belitung failed to satisfy the internal consistency criterion. To address this heterogeneity, the model was re-estimated separately for West Sumatera and Kepulauan Bangka Belitung and for the remaining provinces. The final results show that all districts/cities achieved RSE ≤ 25% and met internal consistency requirements, indicating that the proposed approach improves the precision and reliability of district-level child poverty estimates across Sumatera Island.

Copyrights © 2025






Journal Info

Abbrev

JMST

Publisher

Subject

Agriculture, Biological Sciences & Forestry Mathematics Other

Description

Merupakan media informasi dan komunikasi para praktisi, peneliti, dan akademisi yang berkecimpung dan menaruh minat serta perhatian pada pengembangan Matematika, ilmu pengetahuan dan teknologi. Diterbitkan oleh Lembaga Penelitian dan Pengabdian kepada Masyarakat, Universitas ...