Pemodelan Indeks Pembangunan Manusia (IPM) di Indonesia Menggunakan Geographically Weighted Ordinal Logistic Regression (GWOLR)

Fariz, Fariz (2022) Pemodelan Indeks Pembangunan Manusia (IPM) di Indonesia Menggunakan Geographically Weighted Ordinal Logistic Regression (GWOLR). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 06211740000097-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2024. Download (2MB) | Request a copy

Abstract

Indeks Pembangunan Manusia (IPM) sebagai angka yang mengukur capaian pembangunan manusia berdasarkan tiga dimensi dasar yaitu umur panjang dan hidup sehat, pengetahuan, dan standar hidup layak. IPM dipengaruhi oleh beberapa faktor; yaitu angka kesakitan, rasio sekolah dan murid SMA, rasio guru dan murid SMA, persentase penduduk miskin, tingkat partisipasi angkatan kerja, dan kepadatan penduduk. Perbedaan kondisi geografis dalam mendapatkan faktor pembentuk IPM pada Provinsi di Indonesia menjadikan umur panjang dan hidup sehat, pengetahuan, dan standar hidup layak di Indonesia merupakan permasalahan yang spasial. Model GWOLR dapat menjadi solusi untuk model regresi yang koefisiennya bergantung pada lokasi geografis diamatinya data. Maka dari itu digunakan analisis GWOLR pada penelitian ini. Hasil pemodelan GWOLR dengan pembobot gaussian dengan nilai AIC terkecil diperoleh model GWOLR terbaik dengan faktor rasio sekolah dan murid SMA, rasio guru dan murid SMA, persentase penduduk miskin, dan kepadatan penduduk. Hasil pengujian secara serentak model GWOLR terbaik diperoleh hasil bahwa nilai dari likelihood ratio test sebesar 124,276, yang berarti rasio sekolah dan murid SMA, rasio guru dan murid SMA, persentase penduduk miskin, dan kepadatan penduduk mempengaruhi kategori IPM secara bersama-sama. Hasil ketepatan klasifikasi model GWOLR terbaik sebesar 94,12%.
===============================================================================================
Human Development Index (HDI) is a number that measures the achievement of human development based on three basic dimensions, namely a long and healthy life, knowledge, and a decent standard of living. HDI is influenced by several factors; namely the morbidity rate, the ratio of school to high school students, the ratio of teachers to high school students, the percentage of the poor, the labor force participation rate, and population density. Differences in geographical conditions in obtaining HDI forming factors in provinces in Indonesia make long life and healthy life, knowledge, and a decent standard of living in Indonesia a spatial problem. The GWOLR model can be a solution for a regression model whose coefficients depend on the geographic location of the observed data. Therefore, GWOLR analysis was used in this study. The results of GWOLR modeling with gaussian weighting with the smallest AIC value obtained the best GWOLR model with a factor the ratio of school and high school students, the ratio of teachers and high school students, the percentage of poor people, and population density. The results of the simultaneous test of the best GWOLR model showed that the value of the likelihood ratio test was 124,276, which means the ratio of school and high school students, the ratio of teachers and high school students, the percentage of poor people, and population density affect the HDI category together. The result of the best GWOLR model classification accuracy is 94.12%.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	GWOLR, IPM, Modelling, Provinces, Spatial, Pemodelan, Provinsi, Spasial
Subjects:	G Geography. Anthropology. Recreation > G Geography (General) G Geography. Anthropology. Recreation > G Geography (General) > G70.212 ArcGIS. Geographic information systems. Q Science > Q Science (General) Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Fariz Fariz
Date Deposited:	18 Feb 2022 19:30
Last Modified:	31 Oct 2022 03:03
URI:	http://repository.its.ac.id/id/eprint/94623

Actions (login required)

View Item