Applying housing feature pricing theory, the housing estate's average price in Nanjing was selected as a dependent variable in Sep.2011,and 8 factors affecting housing prices as an independent variable is quantized. Then geographically weighted regression was processed. Compared with the traditional OLS estimation, GWR model has a better fit. At the same time, it has been found that all significant factors influencing housing prices in the spatial distribution are different and the characteristics of this difference are revealed.