APPLICATION OF SPATIAL AUTOREGRESSIVE MODEL ANALYSIS USING BISHOP CONTIGUITY AND SPATIAL ERROR MODEL USING QUEEN CONTIGUITY

Case study : Factors Affecting the Human Development Index on Sulawesi Island in 2022

Abstract

Spatial weighting matrix is very important in describing the relationship between one loca- tion and another in spatial regression. In this study the authors compared the bishop contiguity weighting matrix in the SAR model and queen contiguity in the SEM model on HDI data on the island of Sulawesi. The variables used are HDI, AHH, HLS and per capita expenditure. The pur- pose of this research is to know the best model used in modeling the factors that influence HDI on the island of Sulawesi and provide literature that the selection of weighting matrix in spatial analysis is very influential on the results of the study. The research results show that the AIC value of the SEM model using queen contiguity weighting is 194.7679, while the AIC value of the SAR model using bishop contiguity weighting is 211.0294. Based on the AIC value, it can be concluded that the best modeling obtained is SEM with queen contiguity weighting.