Printer Friendly

The autocovariance of expenditure shares from consumer expenditure survey data.

Jeffrey W. Smith and Mary F. Kokoski, "The Autocovariance of Expenditure Shares From Consumer Expenditure Survey Data," presented at the Allied Social Science Association Annual Meetings, Society of Government Economists, Chicago, IL, December 28 - 30, 1988.

The continuing Consumer Expenditure Survey uses a rotating panel sampling design. As a result of this design, approximately 80 percent of the households sampled in one (calendar) quarter will also be sampled in the following quarter. To the extent that the expenditure patterns of individual households are autocorrelated over time, quarterly estimates of aggregate expenditure shares will exhibit autocovariation. The presence of autocovariation can be shown to create bias in the construction of quarterly chained indexes, and the magnitude of this bias is related to the magnitude of the covariance. The purpose of this paper is to determine this magnitude. The issue is addressed both analytically and empirically.

The analytical expression for the autocovariance is derived in a framework of random sampling from a superpopulation. The effects of sample size are considered, and it is shown that the magnitude of the covariance declines on the order of approximately 1/n. In addition, two relevant features of expenditure share estimates are considered: the fact that each share must lie between zero and 1, and the fact that the shares must sum to unity. It is shown that both of these features imply that the magnitude of the covariance is very low.

The empirical part of the paper looks at the expenditure shares of households that are surveyed in adjacent quarters. For these households, the correlation between expenditure shares for various expenditure categories is calculated. It is found that shares for nondurables tend to exhibit positive autocorrelation and those for durables show negative autocorrelation. Typically, the magnitude of the estimated correlations is found to be small.

Overall, both the analytical and empirical parts of the paper show that, even though household expenditure shares may be autocorrelated, the autocovariance of estimates of aggregate expenditure shares is sufficiently small that it can be safely ignored.
COPYRIGHT 1988 U.S. Bureau of Labor Statistics
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1988 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Consumer Expenditure Survey Conference paper summaries
Author:Smith, Jeffrey W.; Kokoski, Mary F.
Publication:Monthly Labor Review
Date:Aug 1, 1988
Previous Article:Individual consumption within the household: a study of expenditures on clothing.
Next Article:Goods vs. services: from the perspective of consumer spending.

Related Articles
Distribution of consumption examined using aggregate expenditure shares.
Consumer expenditures: results from the Diary and Interview surveys.
Spending patterns of older persons revealed in expenditure survey.
The Consumer Expenditure Survey: quality control by comparative analysis.
BLS to publish quarterly data from Consumer Expenditure Survey.
Consumer expenditure and inequality: a budget components analysis using the Gini coefficient.
Individual consumption within the household: a study of expenditures on clothing.
Goods vs. services: from the perspective of consumer spending.
Expenditures of urban and rural consumers, 1972-73 to 1985.
The CE and the PCE: a comparison: an analysis of a decline in the ratios of aggregate spending for various categories of expenditures from the BLS...

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters