PEMODELAN JUMLAH PENDUDUK MISKIN DI JAWA TENGAH MENGGUNAKAN GEOGRAPHICALLY WEIGHTED REGRESSION (GWR)

Abstract

Poverty is something that is often a measure of the success of a regional head's leadership. It is also the first goal of Sustainable Development Goals (SDG’s) to be alleviated. The right policy is very important to be made for the achievement of sustainable development goals. Geographically Weighted Regression (GWR) modeling is important to be used to develop models in each district / city as a basis for policy makers. The variables used in this study are the number of poor people, Human Development Index (HDI), unemployment rate (TPT), and regency/municipality minimum wage (UMK). The purpose of this study is to determine the factors that influence the number of poor people in each regency/municipality in Central Java. GWR modeling is more effective in describing the number of poor people in regencies/municipalities in Central Java in 2018. This is indicated by the increase in the value of R² and the decrease in the value of Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE).