Hedonic Analysis for Estimation of Condominium Rent Utilizing WEB Information

Information in web is introduced to hedonic analysis for real estate rent. 23 wards in Tokyo are chosen for the data collection areas and rents of condominiums are estimated. To utilize the web site information, variables signifying areal reputation are generated from weblog data set. The results show that the contribution of such information varies from area to area, and that especially in a well reputed areas the contribution tend to become larger. This suggests that such information can be useful for hedonic analysis for well reputed areas, and qualitative aspects such as good townscape and sense of high class can be incorporated with this method to improve the accuracy of the rent/price estimation of real estate.