Geographically Weighted Probit Ordinal Regression Model Estimation

Department of Statistics, Faculty of Science and Data Analytics, Institut Teknologi Sepuluh Nopember Surabaya 60111, Indonesia

^* Corresponding author: This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

Geographically Weighted Probit Ordinal Regression (GWPOR) is a combined method between Geographically Weighted Regression and Probit Ordinal Regression. This study estimates the percentage of poor people using the GWPOR method. In this study, using data from 119 districts/cities on the island of Java, Indonesia. The results show that The research results show that the AICc of the GWPOR model is lower than the AICc of the Ordinal Probit model. This highlights the effectiveness of spatially weighted ordinal modeling in capturing local variations in the determinants of poverty, thereby providing valuable insights for individual regional policymaking. in the GWPOR model. There are 48 districts/cities where the percentage of poor people is influenced by Average Years of Schooling, 50 districts/cities where it is influenced by Average Years of Schooling and Gross Regional Domestic Product (GRDP), 11 districts/cities where it is influenced by Average Years of Schooling and Life Expectancy Rate (LER), 11 districts/cities where it is influenced by Average Years of Schooling, Life Expectancy Rate (LER) and Labor Force Participation Rate (LFPR). The GW-Ordinal Probit model can be a powerful analytical tool to support the formulation of policies that are contextual and more targeted in each region.