The Second-hand Housing Market Analysis And Price Forecast In Jinan

Nowadays,the inflated housing price has always been a hot topic of social concern and discussion.The analysis of the real estate market is also the subject of many scholars.There are many volatility factors that affect the housing prices.These factors are almostly uncertain.As the real estate market is a very complicated and changeable nonlinear system,it is particularly important to analyze the data of second-hand housing market and accurately predict the transaction price of second-hand housing.With the development of the times and the explosive growth of all kinds of information,it is crucial to accurately excavate and further predict the housing price in the complicated housing price information.Most scholars predict the housing price from the perspective of macro economy,such as interest rate and CPI,while there are relatively little research on the prediction and analysis of the one-room-one-price of the building.This article forecasts the price of one room and one price for each house in the secondhand housing market in Jinan.The data selects the information of the second-hand houses that have been sold on the website of the second-hand house intermediary,and the time span is from January 2018 to December 2018.Before predicting the housing price,we need to have a general understanding and grasp of the second-hand housing market in Jinan by using crawled data.With the help of ggplot2 in R language,this paper makes a visual analysis of the housing transaction quantity,housing type,orientation,building age,building area,floor height,total price of housing in each district and unit price of housing in each district,so as to have a more intuitive understanding of the market.The core content of this paper is to accurately predict the housing price.Nine indicators,including area,number of people concerned,apartment type,floor height,total floor number,building area,building age,decoration and equipped with elevators,are selected to study their influence on the housing price.In order to explain that the selected nine indicators are appropriate and correct,a gray correlation analysis is performed,and the calculation results are better.In this paper,a BP neural network model is built based on the second-hand housing transaction data to predict the second-hand housing transaction price.The following conclusions are drawn: the goodness of fit of the model reaches 96.27%.The relative error between the second-hand housing predicted price and the real transaction price fluctuates in [-6,6],and the margin of error is small.Furthermore,it shows that the model is effective in predicting the second-hand housing price in Jinan.Finally,by using the 85 newly climbed second-hand house data,the author predicted the housing price and found that 72.9% of the predicted price was slightly lower than the listed price.In addition to uncontrollable error factors and human factors,this is also related to the intervention of intermediary agencies.