PEMODELAN DATA KEMISKINAN DI PROVINSI SUMATERA BARAT DENGAN METODE GEOGRAPHICALLY WEIGHTED REGRESSION (GWR)
Counting the number of poor have often been modeled as a function of a global regression, which meant that the regression coefficient value applied to all geographic regions. Though this assumption was not always valid because of the differences in geographic locations most likely causing the spatia...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Universitas Diponegoro
2013-06-01
|
| Series: | Media Statistika |
| Online Access: | https://ejournal.undip.ac.id/index.php/media_statistika/article/view/5663 |
| Summary: | Counting the number of poor have often been modeled as a function of a global regression, which meant that the regression coefficient value applied to all geographic regions. Though this assumption was not always valid because of the differences in geographic locations most likely causing the spatial heterogeneity. In case of spatial heterogeneity, the regression parameters would vary spatially, so if the global regression model was applied, would produce an average value of those regression parameters which vary spatially. This study uses the method Geographically Weighted Regression (GWR) to analyze data that contains spatial heterogeneity. In GWR model estimation, the model parameters are obtained by using the Weighted Least Square (WLS) which gives a different weighting in each location. This study discusses the factors that influence the level of poverty in the province of West Sumatra. Suitability test of the model results shows that there is no influence of spatial factors on the level of poverty in the province of West Sumatra. The results shows that there are four variables that are assumed to affect the level of poverty in the province of West Sumatra, they are the variable of floor space, the facility to defecate, ability to pay the cost of health center / clinic and education levels of household head. The four variables have a similar effect in every city and county.
Keywords : Poverty, Spatial Heterogeneity, Geographically Weighted Regression |
|---|---|
| ISSN: | 1979-3693 2477-0647 |