Econometric issues in hedonic price indices: the case of internet service providers

Researchers in hedonic studies frequently encounter the problems of the choice of functional forms, the use of pooled regression using time dummies vs period to period regression, and the unit of measurement of the product. This article examines these issues through the study of Internet service providers in Canada from 1993 to 2000. A series of tests are employed to evaluate the best procedure. We find that the commonly used log-linear equation with period to period regression and hourly rate charged gives a robust result compared with the more flexible translog function. The quality-adjusted price index declines at about 15% per year.