This research aims to model rice consumption in South Sulawesi Province in 2021 by incorporating spatial effects. The methods used are linear regression and Geographically Weighted Regression (GWR). In the GWR model, parameter estimation uses Weighted Least Square (WLS) with Adaptive Gaussian kernel function weighting. The results concluded that the GWR model is better than the linear regression model. Based on the results of the study, there are 3 variables that are thought to affect the rice consumption of districts and cities in South Sulawesi Province, namely GRDP per capita (X1), poverty line (X2), and rice production (X3). The GWR model proved to be able to increase the value and decrease the AIC value. The diversity of the response variable can be explained by the model by 98.22%, the rest is influenced by other variables outside the model. The AIC value obtained in the GWR model is 486,81.
Copyrights © 2023