Spatial panel data model on human development index at Central Java

Human development is one of the ways to create human resources that have the ability to drives the economy. It is become the focus of Indonesia's government attention and listed on several programs implemented at the global and national levels. Human development is measured by one composite index known as the Human Development Index (HDI). HDI is often used as a measure of government success in human development. Human development in Indonesia could vary because it depends on the policies of each region, so in a region, it could give an influence between regencies or cities. In this study, focus of attention is Central Java because HDI in this region has always increased from 2017 to 2019. Therefore, this study aims to analyze the factors that affect HDI on regencies or cities in Central Java from 2017 to 2019. Factors used in this study based on previous research are gross regional domestic product, high school participation rates, population density, labor force participation rates, percentage of poor population, and percentage of households with proper sanitation. Based on this study the observed data contains location information, so these observations could be spatially correlated. Method that could handle the issue and also observed in three time periods (years) is spatial panel data model. The spatial panel data model in this study is divided into two, known as spatial lag panel data model and spatial error panel data model. Based on the comparison of two models, it is obtained that the best model could explain the Human Development Index at Central Java from 2017 to 2019 is spatial lag panel data model with all the explanatory variables significant.