|
Activity Number:
|
343
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Business and Economics Statistics Section
|
| Abstract - #309476 |
|
Title:
|
Smoothed Moments Variance Estimator for X-11 Seasonal Adjustment
|
|
Author(s):
|
Daniell Toth*+ and Stuart Scott
|
|
Companies:
|
Bureau of Labor Statistics and Bureau of Labor Statistics
|
|
Address:
|
Office of Survey Methods Research, Washington, DC, 20212-0001,
|
|
Keywords:
|
seasonal adjustment ; X-11 ; time series ; survey error
|
|
Abstract:
|
Pfeffermann's method for estimating variances for X-11 seasonally adjusted series uses the method of moments at a key step. With or without the use of sampling error autocovariances in the method, a decision must be made on the order of an error process. In the former case, negative estimates of error variance can arise in practice. Using X-11 output, this paper models the unknown noise process in order to avoid inappropriate error autocovariances and to try to improve order selection. Comparisons are made to the methods in Pfeffermann (1994) and Pfeffermann and Scott (1997).
|
