|
Activity Number:
|
503
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 10, 2006 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Government Statistics
|
| Abstract - #306014 |
|
Title:
|
Revisions and Other Changes in Official Statistics: Swedish Observations on Concepts and Terminology
|
|
Author(s):
|
Eva Elvers*+
|
|
Companies:
|
Statistics Sweden
|
|
Address:
|
Department for Research and Development, Stockholm, 10451, Sweden
|
|
Keywords:
|
preliminary statistics ; final statistics ; revision ; correction ; recalculation
|
|
Abstract:
|
The term revision is often used broadly for any change in a value of released official statistics. Current Swedish work on a more precise meaning of preliminary statistics has been extended to some related concepts and terms. Different types of changes should be distinguished for the benefit of users and producers. Corrections are necessary but unplanned changes, e.g. when influential mistakes are detected. Revisions are planned actions. Final statistics are the only release or the last one in a sequence. Furthermore, new knowledge or new data sources may motivate new estimation procedures and releases, even after the final statistics. Such statistics are called recalculations. Environment statistics and national accounts are typical examples. Preliminary and final statistics can both be corrected and recalculated. Preliminary statistics but not final statistics can be revised.
|
