ANALISIS FAKTOR-FAKTOR YANG MEMPENGARUHI KEMISKINAN DI JAWA TENGAH MENGGUNAKAN MODEL GALAT SPASIAL

Abstract

Poverty is one of problems in developing country like Indonesia. From year to year, poverty in Central Java has decreased. This study is aimed to know the poverty model in Central Java by using Spatial Error Model. This research uses data from the number of poor people in Central Java in 2012. Spatial Error Model is a spatial method that showed spatial autocorrelation in the error. In Spatial Error Model, there are spatial dependency effect and spatial heterogenity. The variables that significantly affect the number of poor people in Central Java through Spatial Error Model are the percentage of 10 years old–over population with the highest education is primary school ( X 2 ) and the number of households that have access to reliable drinking water (X 3 ). This Spatial Error Model results R 2 are 75,39% with the AIC are 63,36. It is better than regression model of Ordinary Least Square (OLS) which produces 66,3% of R 2 with AIC are 69,286. It showed the poverty model in Central Java by using Spatial Error Model is better than regression model of Ordinary Least Square (OLS) and in OLS assumption of homoskedasticity not significant. Keywords : Poverty, Regression, Ordinary Least Square, Spastial Error Model