Statistical offices try to match item models when measuring inflation between two periods. However, for product areas with a high turnover of differentiated models, the use of hedonic indexes is more appropriate since they include the prices and quantities of unmatched new and old models. The two main approaches to hedonic indexes are hedonic imputation (HI) indexes and dummy time hedonic (HD) indexes. This study provides a formal analysis of the difference between the two approaches for alternative implementations of an index that uses weighting that is comparable to the weighting used by the Törnqvist superlative index in standard index number theory. This study shows exactly why the results may differ and discusses the issue of choice between these approaches. An illustrative study for desktop PCs is provided. Abstract:Statistical offices try to match item models when measuring inflation between two periods. However, for product areas with a high turnover of differentiated models, the use of hedonic indexes is more appropriate since they include the prices and quantities of unmatched new and old models. The two main approaches to hedonic indexes are hedonic imputation (HI) indexes and dummy time hedonic (HD) indexes. This study provides a formal analysis of the difference between the two approaches for alternative implementations of an index that uses weighting that is comparable to the weighting used by the Törnqvist superlative index in standard index number theory. This study shows exactly why the results may differ and discusses the issue of choice between these approaches. An illustrative study for desktop PCs is provided.

The paper argues for the use of scanner data from EPOS systems for use in the compilation of consumer price indices. A number of methods of calculating micro‐indices from such data are outlined. Scanner data for colour television sets in the U.K. are used as an example. The Tornqvist chained index is used as a benchmark against which alternative formulations, including those based on representative products, can be judged, the errors often being substantial. The paper argues for the use of scanner data, illustrates methods of compiling micro‐indices and points to the potential for serious errors from conventional methods.

No abstract

This Working Paper should not be reported as representing the views of the IMF.

No abstract