Hello Elisa, that looks fine to me. It is maybe useful to add an extra line on your query that sorts the result according to application and application filing date. That will allow you to also easily identify those patent applications for which there were more then one opponent. Each new extra oppo...

You could use any of the methods to generate a "random sample" (which is a strongly debated concept in SQL; how random is random...) But SQL questions are outside the scope of PATSTAT support. Nevertheless, here is an example. SELECT top 5000 tls201_appln.docdb_family_id ,tls201_appln.appl...

When joining tables in relational databases, the data base basically "de-normalizes" the data which results into duplicates. An EP-B1 publication (what you used) is indeed a unique publication. But if you link it to the CPC table, you will get a new row for each CPC code assigned to that a...

The table is based on the application filing year , and the name of the patent owner as currently registered in the EPO register. That does not necessarily mean that the patent was already granted "at the time of the deal." When I look at the timeline, I would suspect that Astrazeneca exec...

And here is the link:
https://data.epo.org/expert-services/index.html

Hello Sanobe, in the first post of the forum, there is an example that closely matches what you are looking for: https://forums.epo.org/sql-queries-to-create-simple-applicant-year-based-table-10588 But if you want to better observe what happens before and after a merger, it would be advisable to use...

Hello JohnVoyore, Your query will not generate any duplicate applications as such, because each application has only 1 record in tls201_appln and can only have 1 title, so you will not have any duplicates. You might of course have patents filed in different patent offices for the same invention. Tho...

Hello alkis.pit, When looking for patents linked to a certain applicant it is better to use the PATSTAT standardised name. (psn_name). The best way to find out which psn_name(s) is the best for your purpose is to run a quick query doing a patent count. Something like this: SELECT psn_name, Count(tls...

Hello Philippe, I am not sure what statistics exactly you would like to have, but in the linked excel sheet you will find data on the allocation and / or distribution of the IPC and CPC codes. (data was extracted from PATSTAT, the SQL code to do so are included in the document.) http://webserv.epo.o...

Hi Zecon, I should have been more precise in my explanation. Very limited bibliographical data on provisional applications is available in PATSTAT because they have been used as a priority filing to establish a priority date. But they were not published. So there is hardly bibliographical data excep...