Skip to content

Statistical processing

Contact info

Prices and Consumption, Economic Statistics
Martin Sædholm Nielsen
+45 39 17 30 05

mne@dst.dk

Get as PDF

Consumer Price Index

The consumer price index is calculated on the basis of 23,000 prices collected from approx. 1,600 shops, companies and institutions throughout Denmark. Most prices are by far collected monthly. The data material received is examined for errors, both by computer (using the so called HB-method) and manually. The different goods and services, which are included in the consumer price index, are first grouped according to approx. 500 elementary aggregates for which elementary aggregate indices are calculated. The elementary aggregate indices are weighted together into sub-indices that are in turn aggregated into the total consumer price index. In calculating a price index it is assumed that the baskets of goods that are compared are identical, also with respect to the quality of the goods. Mainly indirect quality adjustment methods are being applied in the consumer price index in connection with changes in the sample.

Source data

The consumer price index is calculated on the basis of 23,000 prices collected from approx. 1,600 shops, companies and institutions throughout Denmark. The prices are collected by questionnaires or by means of price collectors who visit the individual shops. Prices on food and beverages and household articles are to a large extent covered by the use of scanner data from supermarket chains. Prices are also obtained via the internet. Most prices are by far collected monthly. For goods and services, where prices typically change less frequently, prices are collected more rarely, for instance quarterly or biannually. The index weights for the detailed indices (elementary aggregate indices) are calculated on the basis of data from the national accounts on final consumption expenditure of households in Denmark, supplemented by detailed information from the Household Budget Survey.

Frequency of data collection

Most prices are by far collected monthly. For goods and services, where prices typically change less frequently, prices are collected more rarely, for instance quarterly or biannually.

Data collection

For clothing etc., prices are collected by price collectors who visit the individual shops. Food and beverages and household items are to a large extent covered by the use of scanner data from supermarket chains. For the remaining groups of goods and services, prices are mainly obtained from the shops via mailed forms with the information requested. Finally, Statistics Denmark obtains information on prices concerning a number of selected goods and services by telephone or via the Internet.

Data validation

The data material received is examined for errors, both by computer (using the so called HB-method) and manually.

Data compilation

The different goods and services, which are included in the consumer price index, are first grouped according to approx. 500 elementary aggregates for which elementary aggregate indices are calculated. The elementary aggregate indices are mainly calculated as geometric indices. The elementary aggregate indices are weighted together into sub-indices that are in turn aggregated into the total consumer price index.

Adjustment

In calculating a price index it is assumed that the baskets of goods that are compared are identical, also with respect to the quality of the goods. Consequently, in the case of changes in quality the prices should, in principle, be adjusted for this. Mainly indirect quality adjustment methods are being applied in the consumer price index in connection with changes in the sample. This means that the quality difference between a good leaving and entering the sample is not calculated directly. The price development of comparable goods in the sample is instead calculated, and it is assumed that the price development between the good leaving and entering the sample is equal to the calculated. A remaining price difference between the good leaving and entering the sample is implicitly assumed to be due to a quality difference and is not included in the calculation of the price index.

As the value of the actual changes in quality is not known, it is naturally difficult to calculate estimates for a possible bias, due to the chosen methods of quality adjustment.