Study On The Price Of Commercial Housing Based On Chaotic Neural Network

Nearly forty years since the reform and opening up,China has a greater degree of reform in state-owned enterprises,education and other negative side;the most important is the reform of the economic system.The economy and market of our country has achieved rapid development,and housing reform is implementing,the state will be economic development focus in the real estate market and the real estate industry become a pillar industry of our country,so from the government to people are concern about the price of real estate deeply.Whether the real estate industry can get healthy and orderly development,it is very important to the development of our country's economy.Therefore,according to prediction results which is to forecast the price of real estate by the effective and accurate prediction method,it is to make plan and decision-making by the government,and to provide positive and effective role in guiding by consumer's investment decision.Based on the previous research results,this paper is using the qualitative and quantitative method which combines of theory and practice.This paper uses the chaotic neural network model to analysis the price movements on commercial housing more rationally.It is to find an effective prediction theory,and expand the research field of chaos neural network model.This paper analysis and summary on the real estate price prediction research At home and abroad,it is to find the innovation points of this article.It is to elaborate the basic theory of the real estate forecast,the characteristics of the real estate market and the real estate market analysis of the impact of factors.It is to find its unique features in other commodity markets.And,according to analyzing the chaotic characteristics of the real estate industry,it is to make clear the applicable conditions of the chaotic neural network model.This paper selects a number of factors as alternative indicators,and sorts the importance of these factors by using of SPSS statistical software.By using the phase space reconstruction theory of chaos theory,the one-dimensional time series of commercial housing price is processed,and the time delay and correlation dimension are calculated by using the method of partial complex autocorrelation and G-P algorithm.The correlation dimension is identified as neural network input neurons and the number of nodes.This paper chooses the range of the neural network hidden layer node number,and calculates the network error minimum in this range to determine the hidden layer node number of neural network.Finally,the topological structure of the chaotic neural network model is established.Finally,combined with the specific data of Tianjin City,it trains the establishment of a chaotic neural network model,and analyses the error,to make clear chaotic neural network model to be suitable for the field of commodity housing price.This topic is intended for looking for effective ways to predict commercial housing prices more reasonably,and thus provide a strong reference for administrative decisions,investment decisions of consumers,as well as expand the chaotic neural network the scope of application,enrich its field of application.