Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Prosiding Seminar Nasional Official Statistics
Sintiya Risla Miftaqul Nikmah
Unknown Affiliation
Humanities Computer Science & IT Economics, Econometrics & Finance Social Sciences

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Mengidentifikasi Faktor Prediktif Struktur Ekonomi Daerah Menggunakan Explainable Machine Learning Berbasis Official Statistics Ulfa Siti Nuraini; Dimas Fatkhul Rahman; Sintiya Risla Miftaqul Nikmah; Krisjen Fraulein Hutagalung
Seminar Nasional Official Statistics Vol 2026 No 1 (2026): Seminar Nasional Official Statistics 2026
Publisher : Politeknik Statistika STIS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34123/semnasoffstat.v2026i1.2805

Abstract

Regional economic structures reflect the characteristics of economic activity across regions. This study aims to identify predictive factors that contribute to regional economic structures in Indonesia by modeling the contribution of 17 economic sectors to Regional Gross Domestic Product (RGDP). The data cover 34 provinces from 2015 to 2021, comprising 238 observations. Multi-Output Extreme Gradient Boosting (XGBoost) was used to predict the contribution of each sector, while SHapley Additive exPlanations (SHAP) was used to identify the variables with the greatest predictive contribution. The results show high predictive performance, with R² values for the test data ranging from 0.86 to 0.99, while cross-validation yielded average R² values ranging from 0.80 to 0.94. The SHAP analysis indicates that demographic characteristics, investment, and human development have high predictive contributions, with varying patterns across sectors. These findings suggest that XGBoost and SHAP can be used to interpretatively map the predictive patterns of regional economic structures, without claiming causal relationships or statistical significance.
Co-Authors Dimas Fatkhul Rahman Krisjen Fraulein Hutagalung Ulfa Siti Nuraini