PEMODELAN FIXED EFFECT GEOGRAPHICALLY WEIGHTED PANEL REGRESSION UNTUK INDEKS PEMBANGUNAN MANUSIA DI JAWA TENGAH

Abstract

Human development index is an indicator for assessing the quality of human resources and measure the results of human development. The achievements of the human development index is not enough if conducting observations in each cities in just one particular time, but the observations need to be made in some period of time. The distribution in each cities is also a concern, because the conditions are so diverse that led to their spatial effects. Therefore, it is necessary to study these variables in some time periods that affect human development index taking into account the spatial effects. Statistical methods used to overcome their spatial effects, especially in the problem of spatial heterogeneity in the data type of panel is Geographically Weighted Panel Regression (GWPR). This study focused on the establishment of GWPR model with fixed effects using fixed exponential kernel on the human development index data cities in Central Java in 2010-2015. The results of this study indicate that the fixed effect model GWPR differ significantly on panel data regression model, and the model generated for each location will be different from one another. In addition, cities in Central Java has five groups based on variables that are significant. In the fixed effect model GWPR generates R2 value of 0,9227. Key words: Human Development Index, Panel Data, Spatial Effects, Fixed Effect, Fixed Exponential Kernel, Geographically Weighted Panel Regression, R2.