Should we set a time window when computing generality or originality index?

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: 11
Joined: Tue Jun 12, 2018 6:30 pm

Should we set a time window when computing generality or originality index?

Post by Darwin » Wed Aug 19, 2020 10:51 am

Hello,

Should we set a time window when computing generality index or originality index of a patent?

I suppose we should set a time window, which just like computing the number of citation of each patent (as the earlier application tends to be cited more than the later applications because it exists longer).

But I did not find any paper mentions that. So i confused about it. Is there any thing I ignore here? and should we set a time window?

thanks in advance.


EPO / PATSTAT Support
Posts: 231
Joined: Thu Feb 22, 2007 5:33 pm
Contact:

Re: Should we set a time window when computing generality or originality index?

Post by EPO / PATSTAT Support » Wed Aug 19, 2020 4:01 pm

Hello Darwin,
your reasoning is correct that the purpose of the time frame is to avoid bias by the publications that have been "longer around".
I have seen studies (Harhoff) that use 5 years, we used 3 years in our getting started paper.
Getting started with Patstat_20140401.pdf
(272.98 KiB) Downloaded 15 times
The bigger your time window, the less you can work with the most recent publication - they need to have been around for 5 years. You could maybe now-cast; but I don't think that the final results would make much difference. There is enough citation data in PATSTAT to try out both models ;)

Geert BOEDT
PATSTAT Support Team
EPO - Vienna
patstat @ epo.org


Post Reply