Implementasi Model Geographically Weighted Logistic Regression Semiparametric (GWLRS) untuk Prediksi Indeks Kedalaman Kemiskinan di Provinsi Jawa Tengah

Kadafi, Ikmal Thariq (2026) Implementasi Model Geographically Weighted Logistic Regression Semiparametric (GWLRS) untuk Prediksi Indeks Kedalaman Kemiskinan di Provinsi Jawa Tengah. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) Buku Skripsi_Ikmal Thariq Kadafi (Full)_compressed-5-20.pdf Download (687kB)
	Text (BAB 1) Buku Skripsi_Ikmal Thariq Kadafi-Bab 1.pdf Download (713kB)
	Text (BAB 2) Buku Skripsi_Ikmal Thariq Kadafi-Bab 2.pdf Restricted to Repository staff only until 9 March 2029. Download (890kB)
	Text (BAB 3) Buku Skripsi_Ikmal Thariq Kadafi-Bab 3.pdf Restricted to Repository staff only until 9 March 2029. Download (1MB)
	Text (BAB 4) Buku Skripsi_Ikmal Thariq Kadafi-Bab 4.pdf Restricted to Repository staff only until 9 March 2029. Download (2MB)
	Text (BAB 5) Buku Skripsi_Ikmal Thariq Kadafi-Bab 5.pdf Download (692kB)
	Text (DAFTAR PUSTAKA) Buku Skripsi_Ikmal Thariq Kadafi-Dafpus.pdf Download (359kB)
	Text (LAMPIRAN) Buku Skripsi_Ikmal Thariq Kadafi-Lampiran.pdf Restricted to Repository staff only Download (1MB)

Abstract

Poverty in Central Java Province remains a strategic issue with diverse spatial characteristics. In September 2024, the poverty rate in Central Java reached 9.58%, higher than the national average of 8.57%, with a Poverty Depth Index (P1) of 1.41. Differences in socio-economic conditions across the 35 regencies and cities indicate spatial heterogeneity that requires a location-based modeling approach. The research problem centers on how to model poverty-depth categories more flexibly by simultaneously considering global and local effects. The urgency of this study lies in the need for poverty alleviation policies that are adaptive to regional characteristics. This research employs the Geographically Weighted Logistic Regression Semiparametric (GWLRS) as the main model, using Adaptive Gaussian Kernel and Queen Contiguity spatial weights. The analyzed variables include Dependency Ratio, District/City Minimum Wage (UMK), Open Unemployment Rate (TPT), number of industries, livable housing, and access to adequate sanitation in 2024. The evaluation results show that the GWLRS Adaptive Gaussian model achieves an accuracy of 0.9143 with a precision of 0.8824. Meanwhile, the GWLRS Queen Contiguity model records an accuracy of 0.8857 and a precision of 0.8333. These findings demonstrate that the semiparametric approach with the Adaptive Gaussian Kernel provides strong classification performance and effectively captures both global and local spatial variations. In conclusion, GWLRS, particularly with the Adaptive Gaussian Kernel, is an effective and flexible approach for modeling spatially-based poverty depth in Central Java in 2024.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Trimono, Trimono	NIDN0008099501	trimono.stat@upnjatim.ac.id
Thesis advisor	Wara, Shindi Shella May	NIP199605182024062003	shindi.shella.fasilkom@upnjatim.ac.id

Subjects:

Q Science > QA Mathematics

Divisions:

Faculty of Computer Science > Departemen of Data Science

Depositing User:

Ikmal Thariq Kadafi

Date Deposited:

09 Mar 2026 07:45

Last Modified:

09 Mar 2026 07:45

URI:

https://repository.upnjatim.ac.id/id/eprint/50181