Calculation of Correspondence Tables & HS Combined Classification

599 views
Skip to first unread message

lasp...@gmail.com

unread,
Jul 8, 2014, 9:53:21 AM7/8/14
to wits...@googlegroups.com
Dear WITS team,

I have two technical/ methodological questions concerning COMTRADE in WITS.

First, how are the correspondence tables used in WITS calculated?
I am espceially interested in the detailed conversion procedures from  HS 6-digit 2012, HS 2007 and HS 2002 to HS 1996 and the conversion procedures from HS to ISIC in general.
In particular, I would like to know,
- how you achieve a consistent concordance between two years (e.g. between HS 6-digits classification in 2001 and 2002)?
- how you achieve a consistent concordance over time (e.g. HS 6-digits classification for 1988/1996 through 2012/13)?
- how you achieve a consistent concordance between two classifications over time (e.g. ISIC and HS for 1996 through 2012/13)?

Or, where do I find detailed information on the methodology used or principles applied in calculating the product concordance tables in http://wits.worldbank.org/product_concordance.html ?


Second, could you explain the HS combined classification further?
In the User-Manual it reads:

A given 6-digit level product may correspond to different product codes in the various HS
versions. Therefore, if dealing with several reporters (possibly reporting in different HS
versions) or several years (the same country may report in different HS versions from year to
year), make sure the selected code is common to all HS versions ...
Q1: How can I make sure whether a selected code is common to all HS versions? And is it sufficient to choose e.g. HS 1996 to ensure that the downloaded Comtrade data are comparable over time? / aka: does WITS perform the correct transformation automatically once I have chosen the HS 1996 classification as derived classification? OR is the HS combined actually some common classification system for the HS 6-digit across time?

Thank you very much in advance for further information, links and helpful comments:)

Best regards,
Lisa Aspalter


pcst...@gmail.com

unread,
Aug 13, 2015, 1:58:14 PM8/13/15
to witsforum, lasp...@gmail.com
If you are still interested in figuring this out, I think this may help. 

http://unstats.un.org/unsd/trade/conversions/HS%20Correlation%20and%20Conversion%20tables.htm

At the bottom there are links to documents that discuss conversion methodology.
Reply all
Reply to author
Forward
0 new messages