The EKS method is an index-number technique for comparing price levels across several countries at once. Named after the statisticians Éltető, Köves and Szulc, it is used to compute purchasing power parities (PPPs) in the Eurostat-OECD programme and, in adapted form, in the International Comparison Program.
Bilateral Fisher indices between pairs of countries are not transitive: comparing A with C directly can give a different answer from going through B. EKS fixes this by taking, for each pair and , the geometric mean of all direct and indirect comparisons through the countries in the set: , where is a Fisher price index. The result is transitive and stays as close as possible to the original bilateral figures, and every country is treated symmetrically. The main alternative, the Geary-Khamis method, instead values quantities at average international prices, which makes results additive but gives larger economies more weight. Wider properties of these formulas are covered under index number bias and tests.
Ref 1100001781