Hedonic Methods

This chapter discusses hedonic methods, which involve applying hedonic regressions to data on the attributes of goods and their prices and using the regression results to form price indexes. Here, attributes explicitly measure the “quality” of the goods (rather than indirectly controlling for quality by tracking identical goods, as in the matched-model method). In this approach, price indexes that hold quality constant are obtained either directly from the hedonic coefficients (dummy-variable approach) or from combining regression estimates with price index formulas (imputation methods).