The simulation non-linear model should be used to carry on the forecast to the non-linear tendency real estate price index.The real estate price index is forecasted by applying BP neural network,but the precision and the restraining speed is all not very ideal.It mainly is because BP neural network itself has the flaw.In order to overcome the flaw of BP neural network,the real estate price index to carry was forecasted by using the wavelet neural network unifying the wavelet transformation and the BP neural network in this paper.The results were compared with the BP network forecast result.Finally it was discovered that forecast using wavelet neural network could achieve very good effect.