住宅价格指数编制研究进展与方法适用性分析

Research Progress and Method Applicability Analysis on the Compilation of Housing Price Index

准确监测住宅价格的波动非常重要,当前的住宅价格指数编制方法存在优化空间,新开发的编制方法也没有完美的,须考虑国情并分场合地使用。本文利用文献计量分析等方法对住宅价格指数编制方法的研究进展进行分析,发现hedonic方法仍是最活跃的研究领域,混合方法是我国近年来被研究最活跃的方法;另外,机器学习开始逐渐流行于房价评估中,为进一步运用于房价指数构建提供了路径,尤其是可以运用于插补法。同时,更多的学者认为,在分层法中采用中位数而非算术平均数降维更合理。本文根据方法的实施特点和优缺点,对方法的适用性进行分析,避免方法使用者及数据运用者误用或误读;考虑研究趋势和实际问题,对分层法、插补法和住宅价格指数的运用场景进行展望。

Accurately monitoring fluctuations in residential prices is crucial.There is room for optimization in the current methodology used to compile residential price indices,and the newly developed methodologies are not perfect either.They need to be used with consideration for the national context and specific circumstances.In this paper,the research progress of residential price index compilation methods is analyzed using methods such as bibliometric analysis.It is found that the hedonic method remains the most active research area,and mixed methods have been the most actively studied approach in recent years in China.Additionally,machine learning is gradually gaining popularity in housing price assessment,providing a pathway for its further application in constructing price indices,especially in imputation methods.Furthermore,more scholars believe that using the median instead of the arithmetic mean for dimensionality reduction is more reasonable in stratification methods.Based on the implementation characteristics also the advantages and disadvantages of these methods,this paper analyzes their applicability to avoid misuse or misinterpretation by users and data analysts.Considering research trends and practical issues,the paper also provides an outlook on stratification methods,imputation methods,and the application scenarios of residential price indices.