中文摘要：

产品质量变化是CPI偏差的主要来源,质量调整问题是CPI统计理论与实践的难点之一。进入大数据时代,CPI质量调整的必要性和可行性都显著上升,常用的一些质量调整方法将更具操作性,而且质量偏差调整的效果也将明显改善。基于大数据的支持,经典的Hedonic方法可采用加权时间―虚拟Hedonic指数和加权时间―产品―虚拟Hedonic指数对质量发生改变的规格品进行价格虚拟,结果更加合理准确,并能保证CPI质量调整的及时性和动态性。目前中国CPI统计尚未引入质量偏差调整,为进一步提高数据质量,应尽快研究和实施CPI质量调整。

英文摘要：

The quality change of products is the main source of CPI bias.The quality bias adjustment is one of the difficulties in the CPI statistical theory and practice.When entering into the era of big data, the necessity and feasibility of the CPI quality adjustment become more obvious.And many approaches of the quality adjustment can be put into effect and get better results.As one of the most important approaches,the Hedonic method will be improved sharply and two new models named the weighted time-dummy Hedonic price index and the weighted time-product-dummy Hedonic price index can be used to estimate the prices of quality-changed products.Both of them will increase the accuracy of the quality adjustment results and promise the timeliness and dynamics.Now,China has not yet carried out the quality bias adj ustment in the CPI compilation.In order to promote the data quality of CPI,China should study on the CPI quality adj ustment and adopt it as soon as possible.