The PGFP codes are the easiest proxy to identify whether or not a granted European Patent entered the national phase. (Legally it is a bit more complicated, but PGFP does the job for nearly all different cases.)

Hello Eduardo, The DOCDB patent family concept is based on a couple of rules from which the main one says: patents having the same priorities belong to the same family. But the fact is that, it happens that the priority information is not correct. This is sometimes reflected by looking at the pdf (p...

See this post: https://forums.epo.org/sql-queries-to-create-simple-applicant-year-based-table-10588 , and adapt the classifications codes to what you need. Be careful with spaces and characters when you write the CPC codes; there should allways be 8 digits before the "/" symbol. In most ca...

Hello khanhhienbi, the two tables are very different, and it is not clear how you would want them "combined". The first one is a list of patents, the second one an aggregated table on total number of filings per year. How would a combined table look like ? About the queries, not sure why y...

Such query would generate more then 2 million records, even when only taking the "first applicant" in account for those applications where there are multiple applicants. Here is a query that gives you the results for application filed in 2015. The free trial lets you only download 10.000 r...

Family-family citations count is a pre-aggregated value in the tls201_appn table with the name: nb_citing_docdb_fam . So you don't need to do any calculations. This is the query that shows you the number for the same application as above. SELECT app.appln_auth, app.appln_nr ,app.appln_kind, app.appl...

Keep in mind that your query ONLY looks at forward citations linked to the publication of the grant - the so called B document-. Examiner praxis is to rather cite the earliest publication, which is the A document (or even a priority filing). Researchers who use citations to measure for example techn...

Hello Muyang, as far as I can see, your query is correct, apart from a mistake in one of the ipc_class_symbol and a bracket that is not needed. Observe that I removed the embedded wild card in the classification codes;it can lead to errors because of the way the IPC classification hierarchy is forme...

Dear PATSTAT subscriber, The 2023 Autumn edition has been tested, found sound, and is now ready for download. The bulk data extraction was done end of July 2023, so for the major patent authorities, patents published until the end of 2022 should be complete. PATSTAT Online databases have not yet bee...

I assume you are using a corrupt data set, or maybe a PATSTAT extraction where that data was not included due to the download criteria.