Research On The Influence Factors Of Housing Price Based On SOM-PCR Model

The real estate market of China ushered in a violent climax during the past 2015 and 2016 years,especially the real estate market of those second and third-tier capital cities.The flourishing development of the real estate industry not only brought the growth of the social economy,but also a series of problems.Due to the various factors affecting the real estate prices,the real estate market in each region is diverse.At the same time,along with the rapid development of data mining and machine learning,machine learning algorithms are widely used in all walks of life.The real estate industry has already become one of the most important applications in data mining.The domestic and foreign scholars pay close attention to the research on the prices influencing factors,which is also the research topic of this paper.In this paper,two clustering SOM neural network models were established to obtain clustering results using the new housing price index monthly data of 23 provincial capital cities.The city agglomeration within significantly changes in housing prices was determined according to the comparison result,and the typical city Hefei was selected as the research object.Next,the principal component regression model was established,including foreign economic trade,industry,energy,services,and investment.The linear regression model were built using variables and the new housing price index data.Finally,four main factors resulting in the rising housing prices were: real estate investment,service industry profitability,import and export trade,asset liability ratio.The last,this paper put forward the relevant policy recommendations on the basis of empirical analysis,hoping to play a reference role in stabilizing the real estate market in Hefei and other cities.