The spatial distribution cause analysis of theft crime rate based on GWR model

Based on the geographically weighted regression (GWR) model, carries a cause analysis for the spatial distribution of theft crime rate. There is a negative correlation between population density and theft crime rate, but the correlation is not significant; the theft crime rate has a positive correlation with road network and police intensity, and the correlation is significant in most streets; there are negative and positive correlation between theft crime rate and average land price, but negative correlation is in most streets and the correlation is not significant. It shows that many social factors have some influence upon the spatial distribution of theft crime rate, and the GWR model is indeed able to analysis the cause of crime rate spatial distribution.