Notes on the Measurement of Price and Quality Changes

introduction As ECONOMETRICIANS have accumulated experience in analyzing dif-ferent bodies of data, their views of what are the major technical problems associated with a particular study have changed markedly. Early studies put most of the stress on "what is the relevant theory" for a particular piece of data, with "theory " providing such broad statements as "quantity purchased should be related to price and income. " The next big step came when we started asking, "What is the relevant variable? " Is consumption related to measured income in the same period, to measured income in the previous period, or to some more elaborate but also rather intangible concept of expected or permanent income? Is the planned output of wheat related to current, past, or expected wheat prices? While earlier workers were not unaware of these problems, the general realization of the impor-tance and fruitfulness of such a question owes much to the works o