Go to Statistics Portal


Statistics Directorate    
The country-product-dummy method (CPD) is the multilateral method used by the ICP to obtain transitive PPPs at the basic heading level through regression analysis.

It treats the calculation of PPPs as a matter of statistical inference, an estimation problem rather than an index number problem. The underlying hypothesis is that, apart from random disturbance, the PPPs for individual products within a basic heading are all constant between any given pair of countries.

In other words, it is assumed that the pattern of relative prices of the different products within a given basic heading is the same in all countries. It is also assumed that each country has its own overall price level for the basic heading and it is that which fixes the levels of absolute prices of the products in the basic heading for the country.

By treating the prices observed in the countries for the basic heading as random samples, the PPPs between each pair of countries and the common pattern of relative prices can be estimated using classical least square methods. The method allows sampling errors to be estimated for the PPPs.

Source Publication:
Eurostat, OECD, 2007, Eurostat-OECD Methodological Manual on Purchasing Power Parities, OECD, Paris – Annex VII, Glossary of terms and abbreviations.

Statistical Theme: Prices and purchasing power partities

Created on Thursday, February 6, 2003

Last updated on Thursday, July 12, 2007