Study On The Real Estate Stock Market Early-warning System Of Our Country

Along with the growth of the economy, the stock market in China expands scalegradually. At the same time it is playing an increasingly important role in economy. So weshould attach great importance to the stock market and support the development of it. Theoperation of the stock market is not stable, it often in high low cycles. In China’s stock market,the characteristics of running in the real estate industry in the stock market has highcorrelation with the whole stock market. It runs significantly reflects the runningcharacteristics of the whole stock market, the small changes will appeared in the real estatestock market firstly. Therefore, this paper effective monitoring and early warning of realestate stock market, then we get the result of the study and put foerward advices.First of all, this paper reviews the existing research results at home and abroad includingtheory of early warning and early warning model. It analyzed the contagion mechanism to thereal estate stock market from three aspects, including Economic development in our country,the real estate stock market and the reality of the real estate market. Based on early warningtheory and the reality environment of the real estate stock market in our country, this paperselected11warning indicators from the macroscopic economy, the real estate stock marketand real estate market to build the index system in order to monitor the operation of real estatestock market in our country.Secondly, this paper used the early warning index system for factor analysis with sampledata for January2006to August2013. It comprehensively evaluated the risk situation ofrecent years, and obtained four main factor score and comprehensive factor score. Then thescores were used to identify early warning threshold and transformed into a vector form as theearly warning model of control output. BP neural network model is established in this paperwith the original variables. The network early warning model could extrapolate four months.January2006to April2006data of network is as training data set, May2012to April2013data is as a network test data set, May2013to August2013is as forecast data. The networkmodel training accurate, and the model through the inspection, this paper forecasts the risk of real estate stock market in our country from September2013to December2013. Then it putsforward some prevention advice.Finally, the author of this article summarized the research and put foreword prospect toearly warning research of the real estate stock market in our country in the future.