Concordance between IPC and SIC or ISIC

[rainbowra]

Hi,

I am wondering whether there is any readily available concordance scheme between International Patent Classification (IPC) and other industry classification (e.g Standard Industry Classification - SIC or ISIC).

I found several papers on the issue, yet the link that seems to contain the mapping file does not exist any more.

Fore example, Johnson, D. K. (2002), “The OECD Technology Concordance (OTC): Patents by Industry of Manufacture and Sector of Use” provides a file via the following (broken, now fixed) link:
https://www.oecd-ilibrary.org/science-a ... 1138670407

Appreciated if you could kindly provide me the correct link for the file or if you would recommend some other concordances schemes to fulfill the matching purpose.

Many thanks.

Lai

[nico.rasters]

You'll find something at http://wiki.epfl.ch/patstat/sector
There are definitely more tables available. See also http://ftp.zew.de/pub/zew-docs/div/inno ... ietsch.pdf

Personally I think a concordance between IPC and SIC is as nonsensical/useless as a concordance between colours and fruit.
It's better to map the applicants to a database such as Compustat and retrieve their SIC codes from there.

[mkracker]

You might be interested in NACE2 as well. NACE2 is the Statistical Classification of Economic Activities in the European Community, Rev. 2 (2008) (in French: Nomenclature statistique des activités économiques dans la Communauté européenne). It serves a similar purpose than the SIC (Standard Industrial Classification) and the NAICS (North American Industry Classification System) and is used by Eurostat.

The 2015 Spring Edition of PATSTAT (planned shipping date: early June) contains the concordance table TLS902_IPC_NACE2 which maps IPC sub classes / IPC main groups to the first 2-4 digits of the hierarchical NACE code. All NACE codes can be found in http://ec.europa.eu/eurostat/ramon/nome ... HIERARCHIC .

The data on which this table is based is provided by Eurostat in co-operation with KU Leven / Belgium.
The data and the methodology to create them is described in https://circabc.europa.eu/sd/a/d1475596 ... 2_2014.pdf . This is the 2014 update of an earlier work done by Ulrich Schmoch.

Using this information, we created in PATSTAT an additional table TLS229_APPLN_NACE2. It holds for every patent the NACE2 code(s) which are derived from the patent’s IPC classification. In case of an assignment to multiple NACE2 codes, a weight is given.

I agree with Nico that using industry information from business databases would be the preferred way. But these databases are neither cheap nor is the matching of applicant names easy. So depending on your need concordance tables can be a sensible alternative.

Another source of information on IPC/SIC concordance is the methodology described by Lybbert and Zolas.
The paper is using PATSTAT patent data.

Lybbert_Zolas.pdf

(622.41 KiB) Downloaded 151 times

[PhDBremen]

Hi,
I have a question regarding the table tls229_appln_nace2. I was wondering how the corresponding weight is constructed? Because as far as I understand, you used the information from table tls902_ipc_nace2 in order to create the table tls229_appln_nace2. However, in table902 every ipc class is just one time assigned to one specific Nace Industry. So my question is, how can it be that one ipc class of one concrete patent is assigned to two different nace2 codes?

Furthermore, how did you evaluate that one patent has a higher weight in one industry than a other patent?

Thank you very much in advance for helping me.

Best regards,
Nils