Pemodelan Kemiskinan Pada Kabupaten/Kota Di Provinsi Jawa Timur Tahun 2015 Dengan Pendekatan Model Regresi Spasial

Abstract

Regression is a technique that can be used for response variables with one or more predictor variables. The purpose of this study is to model poverty in districts / cities in East Java 2015 with a spatial regression approach. In 2015, poverty in East Java has increased compared to the previous year. Therefore it is necessary to identify the factors that affect poverty. The variables used are the percentage of poor population as the response variable and the predictor variables include last elementary school education (X1), school participation rate 7-12 years (X2), informal sector workers (X3), open unemployment rate (X4), household using bamboo walls (X5), and household users of inadequate drinking water sources (X6). The result of this research is the best model to model the percentage of poor people is Spatial Error Model (SEM) with spatial weighting matrix Queen Contiguity and obtained AIC value 191,02 and R2 equal to 77,47%. Factors that have significant effect on the percentage of the poor are school enrollment (X2), informal sector workers (X3), household users of inadequate drinking water sources (X6) and there is an error dependency on one location to another.