Research On Improvement And Application Of Hedonic Price Index

The price index reflects the relative change trend of the price.To get an accurate price index,it is necessary to ensure the "homogenous comparability" of the sample.The price change caused by the quality change should not be reflected in the price index,and needs to be excluded.The hedonic price method is an effective quality adjustment method,which can avoid the deviation caused by subjective factors in the traditional matched model method.The rapid growth of a new generation of information technology has promoted faster product replacement,enhanced functions,and substantially improved quality.It has also made product market competition increasingly fierce,resulting in increasingly frequent fluctuations in the price and sales of products in circulation.Therefore,the theory and practical application of the classic hedonic price index method still need to be improved and innovated to deal with these new situations and problems.At the same time,e-commerce and online payment have developed rapidly in China.The massive transaction data accumulated by e-commerce platforms which provides an opportunity to improve the theory and practice of classic hedonic prices index.In view of this,this article tries to analyze the theory and method of classic hedonic price index,optimize the classic hedonic price index compilation method systematically based on the big data of the e-commerce platform.First,systematic review of the theoretical basis of the hedonic price method composed of theories of Lancaster,Rosen and Diewert,summary and comment the principle,advantages,disadvantages and application of the classic hedonic price method(including: the time dummy variable method,the direct hedonic method,the hedonic price imputation method and the hedonic quality adjustment method)in detail,and point out the shortcomings of the classic hedonic price method(including: Insufficient data sources,insufficient model setting,lack of weights and chain drift)in the new economic era,lay a theoretical and methodological foundation for further optimizing the classic hedonic price method.Secondly,we put forward a comprehensive and systematic method of optimizing the hedonic price index to solve the multiple problems.For the lack of weights and insufficient model setting of the classic hedonic price model,a weighted-ADL hedonic price imputation method was proposed,and the RYGEKS price index with the perfectly matched data was used to resolve the chain drift.Combine the direct hedonic method and the hedonic price imputation method to deal with the new features of the specification products effectively.From the application,compiled hedonic price indexes of typical ICT products such as smartphones,laptops,desktops,tablets,USB drives and mobile hard drives based on the classic hedonic price odel,weighted hedonic price imputation method and weighted-ADL hedonic price imputation method separately based on the big data of e-commerce platforms.By compiling the matched data price index in adjacent periods,measure and separate the deviation of the traditional price index caused by the lagging sample rotation and insufficient quality adjustment.At last,the lack of product quality adjustment is still the main factor causing the upward deviation of the official price index,and lead to an underestimation of real GDP,which will affect the accurate judgment of actual economic growth and productivity.We compile the ICT equipment consumer hedonic price index based on the existing samples,estimate the ICT equipment investment hedonic price index based on the existing samples and the cross-border transfer method separately,adjust the deviation of the traditional ICT equipment price index caused by insufficient quality adjustments.We use the sensitivity analysis for growth simulations model to analyze the potential impact of the deviation quantitatively caused by the input-output composition of China's ICT equipment and the traditional price index on the real GDP growth rate.The research shows:(1)The average goodness of fit of weighted-ADL hedonic price imputation method is significantly higher than before the improvement,the accuracy and explanatory of the improved model have been enhanced significantly.(2)Through the measurement of ICT product hedonic price index,the weighted-ADL hedonic price imputation method corrects the deviation caused by insufficient model setting and lack of weights effectively.The monthly Laspeyres index and Paasche index are prone to severe chain drift,the Fisher index alleviates chain drift to a certain extent,the RYGEKS index can resolve chain drift effectively.The traditional price indexes of all types of ICT product have obvious upward deviations due to lagging sample rotation and insufficient quality adjustment.(3)Due to insufficient quality adjustments,the average annual decline in the price index of ICT equipment is underestimated by about 10 to 12 percentage points,resulting in the underestimation of real GDP growth rate by 0.34 to0.41 percentage points.