Estimation of Housing Prices by Fuzzy Regression and Artificial Neural Network

Abstract

Changes in housing prices concern both individuals and government since they have substantial influence on the socio-economic conditions. Valuations of housing are necessary in order to assess the benefit and liabilities in housing section. The housing price in Iran is based on eight economic indices. The study of trends in housing price has been made by considering the related seasonal data from 16 years ago and using the techniques of Artificial Neural Network Back propagation (ANN-Back propagation) and Fuzzy regression. The results of our experiments indicate that the estimation error (Mean Absolute Percentage Error, “MAPE”) in the ANN-Back propagation technique is less than that in Fuzzy regression. It can be shown, by comparing the estimated housing prices by applying the ANN technique with the observed ones, that the ANN technique has favorably estimated the trends in the changes of housing prices.