ELUCIDATING THE SELF-FORMATION OF LOW-INCOME SOCIO-ECONOMICÂ’ CHARACTER USING GEOSPATIAL ANALYSIS MODELLING
This study developed an alternative method for the local analysis of relationships between low-income socio-economics’ character among the various local modelling approaches, Geographical Weighted Regression (GWR). The complexity originates from the integration of spatially and temporally varying factors underlying the interpretation of socio-economic environmental elements. In addition to spatial autocorrelation and spatial nonstationarity exist widely in Geospatial analysis processes, which are incorporated with Ordinary Least Square (OLS) model. The result found GWR models has achieved better performance than the global OLS model, which the individual data on averages are calculated and explained the locational information and link problematic structure on a map. The techniques are applied and generated the exploratory spatial data analysis, to analyse the spatially varying relationships of low-income socio-economic indicators across low-income settlement point’sdatabase of Bangkok, Thailand. Thus, the self-formation of spatial analysis is characterized and explored scale effect on low-income settlement approaches by the location of socio-economic environmental features. It is possible to offer useful alternatives this novel idea joint application of urban planning and policy decision-making when assessing GWR model.