Applied research on real estate price prediction by the neural network

It is an insistent demand by current real estate industry to establish an easy-operate and logical scientific prediction model. But the real estate price is a chronological sequence with a particular statistic relationship which is difficult to be expressed by a predetermined function or equation. And this character makes it difficult to predict the real estate price. However, the neural network can resolve the problem effectively; Moreover, it can reflect the time variability of real estate price. Hereby this essay gave a real estate price prediction methodology based on BP neural network and Elman neural network, and approved that these two methodologies have a good accuracy. Furthermore Elman neural network can forecast more accurate and constringency faster. This kind of character can has a good effect to forecast the price of real estate.