The Formation Mechanism of House Price Differences in China’s Urban Area in the Same City—Based on Lasso Bayesian Model Averaging Method

Abstract

With the development of economy and the expansion of urban scale, the variation of housing prices within Chinese cities has gradually become significant. Based on the housing prices of 79 districts in Chengdu, this paper analyzes the reasons for such differences. With an index system of the price difference of locational housing, a semi-log linear regression Lasso Bayesian average model is constructed, which can realize variable selection while estimating coefficients. Empirical evidence shows that the inequality of social-economic resources and the imbalance of housing supply and demand in the area are important internal factors that lead to the difference in housing prices between areas within the city; the related housing prices have an extremely significant positive impact, showing that the mobility of the place of purchase and the contagion of the house price are the main reasons for housing price; in addition, influential factors have strong positive interaction effects.