Estate Market Forecast Using Artificial Neural Networks

Abstract

Artificial neural networks are widely used in business disciplines. The objective of this study is to provide independent real estate market forecasts on home prices using artificial neural networks. The Cascade Forward Back Propagation (CFBP) neural network is used to forecast house price, based on selected 13 parameters which are considered as forecast variables. The results of applying the CFBP neural networks methodology to forecast house price based upon selected parameters show abilities of the network to learn the patterns. In all cases, the percent correctly forecast in the simulation sample is above 94 percent. Empirical results support the potential of artificial neural network on house price forecast. CFBP neural networks are successfully used model for prediction, classification and forecasting.
Keywords: Cascade Forward Back Propagation (CFBP), Artificial Neural Network, Business Intelligence.