a country's innovation ability per year by using the number of successfully patent applications (at family level)

Here you can post your opinions, ask questions and share experiences on the PATSTAT product line. Please always indicate the PATSTAT edition (e.g. 2015 Autumn Edition) and the database (e.g. PATSTAT Online, MySQL, MS SQL Server, ...) you are using.
Post Reply

Darwin
Posts: 12
Joined: Tue Jun 12, 2018 6:30 pm

a country's innovation ability per year by using the number of successfully patent applications (at family level)

Post by Darwin » Tue May 12, 2020 2:29 am

Hi,

Could you please give me some suggestions to measure a country's innovation ability per year by using the number of successfully patent applications (at family level)?

I do the following things,
(i) identify the first time an invention is patented and call it the “original patent”,
(ii) record the country of the invention using the Patent Authority that accepts the application of the original patent,
(iii) date patents using the application year of the original patent as the application date is closer to the actual date of innovation and
(iv) focus on utility patents only.

Can I do the (ii) or should I use "the country of residence of its primary assignee (i.e., owner) as the country of the invention"?

if I use the (ii), when a UK company submit an application to US patent authority and finally succeed, it will be considered as US innovation ability.

if I use "the country of residence of its primary assignee",
1) I will have to exclude the applicants where person_ctry_code =' '
2) if two UK companies submit an application for a same invention, should I measure it as one application for UK or two?

Could you please give me some suggestions about this?
Thanks in advance.


EPO / PATSTAT Support
Posts: 425
Joined: Thu Feb 22, 2007 5:33 pm
Contact:

Re: a country's innovation ability per year by using the number of successfully patent applications (at family level)

Post by EPO / PATSTAT Support » Thu May 14, 2020 12:43 pm

Hello Darwin,
there are plenty of scientific and policy publications that touch on the topic of measuring innovation through patent analytics.
One of the main sources is the OECD: https://www.oecd.org/site/innovationstr ... rsion.htm OECD has developed various concepts (for example triadic patents), and is very open on their methodologies and patent based indicators. Most of their patent based research and indicators are based on the PATSTAT database. This allows researchers to replicate the approaches and adapt them to their own research.
To answer your questions:
(i) identify the first time an invention is patented and call it the “original patent”,
Many researchers use the first application of a patent family (mostly DocDB family) as an indicator closest to the moment a decision was taken on protecting an invention. Later filings extend the geographical scope of the sought protection.
(ii) record the country of the invention using the Patent Authority that accepts the application of the original patent,
That would be the patent authority of the first priority filing. For PCT and EP applications this might be too broad if you want to assign a "country" to the invention. Many researchers use the country of the inventor. (often weighed)
focus on utility patents only
... The type of the intellectual property right is defined through the IPR_TYPE attribute in tls201_appln. Use PI (Patent of Invention) to include the US Utility Patents (not UM).
primary assignee (i.e., owner) as the country of the invention ?
PATSTAT does not have a concept of primary assignee. The most popular approach is the country where the first filing happened, confirmed by the country of the inventor (or applicant). You can easily compare the results of different models once you have the data agregation set up for one approach.
if I use the (ii), when a UK company submit an application to US patent authority and finally succeed, it will be considered as US innovation ability.
Assuming the first filing for that family is in the US, in this case, the country of the applicant (UK) would not match / confirm the country of the first filing. Maybe the country of the inventor would back it up ? You will need to make a decision model that covers the variations; maybe the applicant has a specific filing strategy, sometimes tax incentives are steering this strategy as well. (Look up the concept of PATENT BOX)
if two UK companies submit an application for a same invention, should I measure it as one application for UK or two
It will be difficult to identify such filings without analysing the claims to see if it is "the same invention". But even if so, I think it is rather irrelevant. From a data aggregation point of view, both applications are valid applications.
PATSTAT Support Team
EPO - Vienna
patstat @ epo.org


Post Reply