Abstract
This article investigates the theoretical and empirical properties of a true cost-of-living index under conditions of changing preferences. A family of true indexes is defined based on the notion of the current utility function. A particular index, based on the previous period’s utility level, is then defined. Given this definition, a true cost-of-living index is computed based on a quadratic expenditure system estimated with quarterly data from 1960-1981. For empirical purposes, changes in preferences are represented by the linear habit formation hypothesis. This index is then compared with Paasche and Laspeyres indexes. The true cost-of-living index grows somewhat faster than either the Paasche or Laspeyres index. It also displays considerably more variability.
Original language | English (US) |
---|---|
Pages (from-to) | 332-335 |
Number of pages | 4 |
Journal | Journal of Business and Economic Statistics |
Volume | 3 |
Issue number | 4 |
DOIs | |
State | Published - Oct 1985 |
All Science Journal Classification (ASJC) codes
- Statistics and Probability
- Social Sciences (miscellaneous)
- Economics and Econometrics
- Statistics, Probability and Uncertainty