Klasifikasi Indeks Ketahanan Pangan Kabupaten/Kota di Indonesia Menggunakan Regresi Logistik Ordinal Spasial dengan Pendekatan Bayesian

Zavira, Hanif Ziva (2026) Klasifikasi Indeks Ketahanan Pangan Kabupaten/Kota di Indonesia Menggunakan Regresi Logistik Ordinal Spasial dengan Pendekatan Bayesian. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Bab 1) 22083010078-bab 1.pdf Download (249kB)
	Text (Bab 2) 22083010078-bab 2.pdf Restricted to Repository staff only until 18 July 2028. Download (674kB)
	Text (Bab 3) 22083010078-bab 3.pdf Restricted to Repository staff only until 18 July 2028. Download (754kB)
	Text (Bab 4) 22083010078-bab 4.pdf Restricted to Repository staff only until 18 July 2028. Download (1MB)
	Text (Bab 5) 22083010078-bab 5.pdf Download (219kB)
	Text (Daftar Pustaka) 22083010078-daftar pustaka.pdf Download (181kB)
	Text (Lampiran) 22083010078-lampiran.pdf Restricted to Repository staff only Download (227kB)
	Text (Cover) 22083010078-cover.pdf Download (2MB)

Abstract

Food security is an essential aspect of development as it is closely related to the ability of communities to access sufficient, safe, and nutritious food. Differences in social, economic, health, and infrastructure conditions have resulted in varying levels of food security among regencies and municipalities in Indonesia. Such disparities may increase food vulnerability in certain regions, highlighting the need for a classification method that can accurately identify food security levels while accounting for spatial dependence among neighboring areas. This study aims to describe the food security conditions of regencies and municipalities in Indonesia based on the 2024 Food Security and Vulnerability Atlas (FSVA) data, apply Bayesian Ordinal Logistic Regression to classify food security levels, and evaluate the performance of a Bayesian Spatial Ordinal Logistic Regression model. Ordinal Logistic Regression was employed because the response variable, the Food Security Index (FSI), consists of six ordered categories: very vulnerable, vulnerable, moderately vulnerable, moderately secure, secure, and highly secure. The Bayesian approach was adopted to estimate model parameters through posterior distributions, while spatial effects were incorporated using the Conditional Autoregressive (CAR) approach to account for spatial dependence among adjacent regions. The study used data from 514 regencies and municipalities in Indonesia, with predictor variables including the Normative Consumption Production Ratio (NCPR), poverty rate, food expenditure percentage, percentage of households without electricity access, percentage of households without access to clean water, average years of schooling of women, health worker ratio, and stunting prevalence. Parameter estimation was performed using the Integrated Nested Laplace Approximation (INLA) method. The results indicate that the Bayesian Spatial Ordinal Logistic Regression model achieved an accuracy of 92.996%, outperforming the non-spatial model with an accuracy of 92.607%, and produced a DIC value of 331.754 and a WAIC value of 327.342. The developed model was further implemented in a Streamlit-based Graphical User Interface (GUI) providing a national food security summary, FSI classification maps, and a food security prediction simulation.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Hindrayani, Kartika Maulida	NIDN0009099205	kartika.maulida.ds@upnjatim.ac.id
Thesis advisor	Nasrudin, Muhammad	NUPTK4241774675130323	nasrudin.fasilkom@upnjatim.ac.id

Subjects:

Q Science > QA Mathematics

Divisions:

Faculty of Computer Science > Departemen of Data Science

Depositing User:

Hanif Ziva Zavira

Date Deposited:

20 Jul 2026 06:47

Last Modified:

20 Jul 2026 06:47

URI:

https://repository.upnjatim.ac.id/id/eprint/56221