Pemodelan Indeks Pembangunan Manusia Nusa Tenggara Barat Menggunakan Geographically Weighted Regression

Abstract

The Human Development Index (HDI) is an indicator for measuring the level of social and economic development of a country or region. The reality is that a local-based model of autonomy is often needed because of the spatial heterogeneity that can occur due to the territory's geographical, social, cultural, or other conditions. This research aims to find spatial effects affecting HDI in West Nusa Tenggara Province. A method that can be used to accommodate is Geographically Weighter Regression (GWR). GWR analysis is the development of multiple linear regression analysis that can address territorial diversity/spatial heterogeneity so as to produce different models and predictions of parameters for each observation region. The modeling was carried out using the Gaussian Kernel Adaptive spatial weigher with an optimal bandwidth value of 27,1227 and a minimum CV value of 5,2927. The GWR model modeling resulted in 10 models for each observation location and showed that life expectancy variables, school expectance, per capita income, and average school-age significantly influenced the IPM in the West Southeast Nusa Province in 2022 with an R2 of 99.92% and a minimum AIC value of -10,0281.