Value Data and the Fisher Index

In this paper we show how to use value data (price times quantity) to construct Fisher price and quantity indexes. In particular, we think of revenue and expenditure data. This model extends the work of Cross and F?re, who showed how to recover relative prices from value data with no explicit price or quantity information. We examine the accuracy of our model over a range of price changes, firm sample sizes, and response variation, in a Monte Carlo experiment in which firms respond to price changes with error. The model outperforms it component indexes with accuracy levels that increase with response variation.