Activity Number:
|
16
|
Type:
|
Contributed
|
Date/Time:
|
Sunday, August 11, 2002 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Section on Survey Research Methods*
|
Abstract - #300936 |
Title:
|
Assessing the Impact of Imputation on the Sampling Variance of the U.S. Consumer Price Index
|
Author(s):
|
Sylvia Leaver*+ and William Larson
|
Affiliation(s):
|
U.S. Bureau of Labor Statistics and U.S. Bureau of Labor Statistics
|
Address:
|
2 Massachusetts Ave., N.E., Room 3655, Washington, District of Columbia, 20212, USA
|
Keywords:
|
Stratified random groups ; Complex multi-stage design
|
Abstract:
|
Imputation of basic-level price change, due primarily to sample attrition and item unavailability, occurs frequently in price index production. Though current variance estimation procedures do take basic-level imputations into account, it is of interest to assess the contribution to total variance that these imputations represent. In this paper, we present estimates of the sampling variance of price change for the U. S. Consumer Price Index which exclude the additional component of variation attributable to basic-level imputation . Variances are presented for one- and twelve-month price change lags, using a stratified random groups methodology. Estimates are contrasted with production variance estimates for the same series.
|