Analysis
Results
All (2)
All (2) ((2 results))
- 1. Televisions: Quality Changes and Scanner Data ArchivedArticles and reports: 62F0014M2001014Geography: CanadaDescription: This paper is the first in a series of reports examining the possible use of scanner data for constructing price indexes. This case study focuses on televisions and compares their price behaviour taken from current surveying methods with alternative measures obtained from massaging electronic data records on all sales by a retailer over a comparable period. Examination of the price index history for televisions shows that the recognition and adjustment for quality change in the sample and the impact of shifts in purchasing patterns have similar impact on the index numbers. The advantages of scanner data - that they record actual sales and current purchasing patterns - have to be set against the difficulty of recognising quality change. This analysis shows that while there are substantial gains from using scanner data in monitoring and adjusting for purchasing pattern changes, it is difficult to account for quality changes without micro-editing the data. The scanner data set raises statistical issues, largely questions of what aggregation across time, outlets and products should be done, that have to be answered before using it in index estimation. Future analysis will be aimed at resolving these issues.Release date: 2001-06-01
- Articles and reports: 62F0014M1998012Geography: CanadaDescription: This paper examines the methods of adjustment for quality change made in the Canadian Consumer Price Index for the period 1989 to 1994. It finds that in most cases the current Canadian practice ensures that the replacement of one commodity by another, one variety of a commodity by another, or one outlet by another, has no impact on the overall index. The main exceptions to this result occur when replacing varieties of commodities that are purchased only occasionally, and a judgement is made that the quality ratio between the old and new variety is not the same as the ratio of their prices. In these cases there is an impact on the index, up or down, depending on whether the change in price reported is higher or lower than the change in quality. From the experience of the CPI in these six years there has been a correlation between the price ratio of a variety and its replacement and the index movement that derives from the judgement. The direction and size of the impact on the index depends largely on whether an item is replaced with a higher or lower priced item. For these reasons, the paper argues that more attention should be paid to ensuring that the item selection is more representative of current sales than has traditionally been the case.Release date: 1999-05-13
Articles and reports (2)
Articles and reports (2) ((2 results))
- 1. Televisions: Quality Changes and Scanner Data ArchivedArticles and reports: 62F0014M2001014Geography: CanadaDescription: This paper is the first in a series of reports examining the possible use of scanner data for constructing price indexes. This case study focuses on televisions and compares their price behaviour taken from current surveying methods with alternative measures obtained from massaging electronic data records on all sales by a retailer over a comparable period. Examination of the price index history for televisions shows that the recognition and adjustment for quality change in the sample and the impact of shifts in purchasing patterns have similar impact on the index numbers. The advantages of scanner data - that they record actual sales and current purchasing patterns - have to be set against the difficulty of recognising quality change. This analysis shows that while there are substantial gains from using scanner data in monitoring and adjusting for purchasing pattern changes, it is difficult to account for quality changes without micro-editing the data. The scanner data set raises statistical issues, largely questions of what aggregation across time, outlets and products should be done, that have to be answered before using it in index estimation. Future analysis will be aimed at resolving these issues.Release date: 2001-06-01
- Articles and reports: 62F0014M1998012Geography: CanadaDescription: This paper examines the methods of adjustment for quality change made in the Canadian Consumer Price Index for the period 1989 to 1994. It finds that in most cases the current Canadian practice ensures that the replacement of one commodity by another, one variety of a commodity by another, or one outlet by another, has no impact on the overall index. The main exceptions to this result occur when replacing varieties of commodities that are purchased only occasionally, and a judgement is made that the quality ratio between the old and new variety is not the same as the ratio of their prices. In these cases there is an impact on the index, up or down, depending on whether the change in price reported is higher or lower than the change in quality. From the experience of the CPI in these six years there has been a correlation between the price ratio of a variety and its replacement and the index movement that derives from the judgement. The direction and size of the impact on the index depends largely on whether an item is replaced with a higher or lower priced item. For these reasons, the paper argues that more attention should be paid to ensuring that the item selection is more representative of current sales than has traditionally been the case.Release date: 1999-05-13