Analysis Of The Influence Factors Of Beijing’s Housing Prices

In recent years, along with the reform of housing system and land usesystem, the rapid development of China’s real estate industry not only improvesthe housing conditions of urban residents, but also become a pillar of China’smacroeconomic development. However, the continuous and fast rise of housingprice together with the sign of real estate bubble is also casting a shadow overthe stable development of China’s economy.Housing problem concerns the national economy and the people’slivelihood. In the face of adverse effect caused by the real estate prices risingtoo fast, government has given a high priority, and various forms of limitedsales, purchase and price controls are also introduced. These policies point to allaspects of the national economy, using administrative means and marketregulation to control housing prices, at the same time to crack down onspeculation. Practice shows that in recent years, some of these measures haveachieved remarkable results, but also contributed to ‘fake marriage’,‘fakedivorce’and some other social issues.In this paper, we use housing prices in Beijing as an example to explore thereal estate price formation mechanism and influencing factors. Which factorsaffect real estate prices and how can the government carry out macro-controlmore effectively and safely is the problem that this article focuses on and strivesto solve.First, this article introduces three real estate price formation mechanism:the administrative instruction type formation mechanism, cost and profitmechanism and market supply and demand mechanism. It also lists population,interest rate, level of economic development, living standards, various kinds ofprice indices, housing construction and sales and other probable factors to affecthousing prices in economic and social development together with their affectingforms. In the process of collecting data, we make further explanation of theconnotation of the these factors.Secondly, after the exploratory data analysis,we use linear regression model to explain the relationship between the data, andthen selecting variables with methods of multiple linear regression such as Akaike information criterion and variance inflation factor to make the modelmore reasonable and reliable. We use seasonal effects models to deal with datalike per capita disposable income of urban households that hold obviousseasonal characteristics, in order to reduce abnormal effects caused by seasonalfluctuation.From the final model, residual plot and prediction curve, we can ensurerationality. However, the model is not completely in line with reality. Maybesome important factors were omitted in the variable selection, or some of thehidden characteristics such asheteroscedasticity have not been taken intoconsideration. These are also problems need to consider in the follow-up study.