Analysis Of The Factors Affecting The Sales Of Commercial Housing Based On Quantile Regression

Since the new century,the real estate industry booming,and people's lives areclosely related.Many scholars believe that the overall trend of the real estate is the national economic development index.Commercial housing as a rigid demand,have been concerned about intimately.In view of the complexity of the real estate industry itself,as well as the linkage between the external data collected,the traditional regression analysis is not able to effectively interpret the fluctuations in home sales.Koenker Roger and Bassett Gilbert proposed the quantile regression modeling method to estimate the linear relationship between a group of quantile regression variables and explanatory variables,they stressed that changes in the conditional quantile regression,volatility difference curve,is to expand the classical least squares mean model based estimation.Quantile regression relaxes the normality assumption of the ordinary least squares estimation for the data of the research object,it is more suitable for the distribution of skewness,spikes or more outliers.With the quantile changes,the regression model can retain more original information.Firstly,this paper introduces the theoretical knowledge and mathematical properties of quantile regression,and compares the differences between the quantile regression and traditional regression.In the part of empirical analysis,I analyzed the data after processing by factor analysis.By means of quantile regression,it is concluded that the factors that influence the sales of commercial houses are different at different levels.Finally,combined with the data,the median regression is more robust than the mean reversion.