|
Activity Number:
|
565
- Time Series in Government and National Statistics
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, July 31, 2019 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Government Statistics Section
|
|
Abstract #306481
|
|
|
Title:
|
Estimating the Variance of Seasonally Adjusted Series of Monthly Statistics Canada Surveys
|
|
Author(s):
|
Francois Verret* and Catalin Dochitoiu
|
|
Companies:
|
Statistics Canada and Statistics Canada
|
|
Keywords:
|
X-12-ARIMA;
X-11 filters;
replication methods;
linearization techniques;
sample surveys
|
|
Abstract:
|
Seasonal adjustment is done using the X-12-ARIMA method at Statistics Canada. Appropriate variance estimates are needed to support inference based on seasonally adjusted estimates. In their absence, cross-sectional quality indicators of precision of the unadjusted series estimates are sometimes provided to data users along with the seasonally adjusted series estimates. Those indicators are often based on the estimated coefficient of variation of the unadjusted estimates. Statistics Canada is currently exploring options for estimating the variance of the seasonally adjusted series for its sub-annual surveys. A method suitable for every statistical program is sought. This paper discusses the methodologies considered and recent progress.
|
Authors who are presenting talks have a * after their name.
