A Bayesian Spatial Ordinal Logistic Regression Approach for Food Security Index Classification in Indonesia
bit-Tech · 10.08.2026 · 9 (1) · 207-217
DOI: 10.32877/bt.v9i1.3829 ↗Kurz gefasst
Fachartikel in der Zeitschrift bit-Tech, online erschienen am 10.08.2026. Verfasst von Hanif Ziva Zavira und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).
Abstract (Originalsprache)
Food security remains a critical issue in Indonesia despite the country’s substantial agricultural potential. Variations in socio-economic conditions, infrastructure, and public health have resulted in unequal Food Security Index (FSI) levels across districts and municipalities. Since the FSI is categorized into ordered levels ranging from highly vulnerable to highly secure, an ordinal modeling approach is required to appropriately capture the ranking structure of the response variable. In addition, the geographical clustering of food security conditions indicates the presence of spatial dependence among neighboring regions, which may not be adequately addressed by conventional non-spatial models. Therefore, this study aims to develop a Bayesian spatial ordinal logistic regression model for classifying FSI categories in Indonesia. The study uses secondary data from the 2024 Food Security and Vulnerability Atlas (FSVA), covering 514 districts and municipalities, with nine explanatory variables representing food availability, accessibility, and utilization dimensions. Spatial dependence is incorporated using a Conditional Autoregressive (CAR) prior, while parameter estimation is performed through the Integrated Nested Laplace Approximation (INLA) method. The results show that the proposed model achieves an accuracy of 72.18% and a tolerance accuracy of 91.83%, indicating that most prediction errors occur only between adjacent categories. Furthermore, the model identifies significant spatial effects and highlights several important variables, including poverty rate, access to clean water, life expectancy, stunting prevalence, and normative consumption ratio, in distinguishing regional food security levels across Indonesia.
Zitieren
APA
Zavira, H. Z., Hindrayani, K. M., & Nasrudin, M. (2026). A Bayesian Spatial Ordinal Logistic Regression Approach for Food Security Index Classification in Indonesia. bit-Tech, 9(1), 207-217. https://doi.org/10.32877/bt.v9i1.3829
Vancouver
Zavira HZ, Hindrayani KM, Nasrudin M. A Bayesian Spatial Ordinal Logistic Regression Approach for Food Security Index Classification in Indonesia. bit-Tech. 2026;9(1):207-217. doi:10.32877/bt.v9i1.3829
BibTeX
@article{Zavira2026,
author = {Zavira, Hanif Ziva and Hindrayani, Kartika Maulida and Nasrudin, Muhammad},
title = {A Bayesian Spatial Ordinal Logistic Regression Approach for Food Security Index Classification in Indonesia},
journal = {bit-Tech},
year = {2026},
volume = {9},
number = {1},
pages = {207--217},
doi = {10.32877/bt.v9i1.3829}
}
EUVIMED