Premium
PROCEDURES FOR UPDATING THE RESULTS OF INTERNATIONAL COMPARISONS
Author(s) -
Szilágyi György
Publication year - 1984
Publication title -
review of income and wealth
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.024
H-Index - 57
eISSN - 1475-4991
pISSN - 0034-6586
DOI - 10.1111/j.1475-4991.1984.tb00483.x
Subject(s) - extrapolation , benchmark (surveying) , interpolation (computer graphics) , econometrics , reliability (semiconductor) , base (topology) , computer science , statistics , economics , mathematics , geography , artificial intelligence , motion (physics) , mathematical analysis , power (physics) , physics , geodesy , quantum mechanics
This article reviews the problems involved in updating the results of international comparisons, in terms of an analytic framework focusing upon the sources of differences between various forms of extrapolation and direct comparisons. The factors identified as important are conservation of prices of the base period and weight inconsistency. The reliability of updating is undoubtedly affected by the length of the period over which the data are extrapolated. A program of regular benchmark comparisons at approximately five‐year intervals with updating for the intervening years is attractive, since it permits checking by forward and backward interpolation. Where there are large deviations, however, averaging is not an acceptable solution.