The NBER U.S. Patent Citations Data File: Lessons, Insights, and Methodological Tools

These data comprise detail information on almost 3 million U.S. patents granted between January 1963 and December 1999, all citations made to these patents between 1975 and 1999 (over 16 million), and a reasonably broad match of patents to Compustat (the data set of all firms traded in the U.S. stock market).

These data are described in detail in

Hall, B. H., A. B. Jaffe, and M. Trajtenberg (2001). "The NBER Patent Citation Data File: Lessons, Insights and Methodological Tools." NBER Working Paper 8498.

ALL USERS OF THESE DATA SHOULD READ THIS PAPER, AND SHOULD CITE IT AS THE SOURCE OF THE DATA

Further documentation on uses of the patent citation data, including the methodology paper and a CD containing the complete dataset itself, is available in the book Patents, Citations and Innovations: A Window on the Knowledge Economy by Adam Jaffe and Manuel Trajtenberg, MIT Press, Cambridge (2002). The book may be ordered from MIT Press. ISBN 0-262-10095-9.

The CUSIP match is based on the 1989 universe of companies

The data are freely available below in two compressed (".zip") formats: SAS transport (.tpt) files and ASCII comma-separated variable (.csv) files. The program read_tpt.sas can be used to convert the .tpt files to native SAS data sets. Lines in the ASCII CSV files are terminated by the newline character "\n". "CSV" stands for comma separated values. All values in the ASCII CSV files are separated by commas. In addition, the character values are enclosed by double quotes. The compression ratio for the compressed files is about 75%. The ".zip" files can be uncompressed with winzip or pkunzip. To check your ability to uncompress these files, download the small file compress.zip. The SAS ".tpt" files are transferable to other formats using software such as Stat/Transfer or DBMS/Copy, and can be used directly by Stata using the fdause command.
To download files in Internet Explorer, right click on them and select "Save Target As...".
Internal users can access the data at /home/data/patents

You will need a major database, statistical program, or programming language to use these files. Most of the datasets are too large to load completely into MS Excel 2000, which has a maximum of 65,536 observations, though Access can be used to read the ASCII datafile. View variable descriptions and observations per file in the "Documentation" column of the table below.
U.S. patent information can also be downloaded or purchased from the United States Patent and Trademark Office, which also has a U.S. to IPC concordance.

To search patents, try Google -> more -> patents or http://www.freepatentsonline.com

For international patent databases check FIZ Karlsruhe, the British Library (Derwent is one Patent Copy Service that delivers patents from the British Library.), the German Patent and Trade Mark Office, Espacenet, Micropat, the French Intellectual Property Institute, the IciMarques database, or the EP-CESPRI database, a database along the lines of the NBER dataset, but for European Patent Office data.
Many of the sources above were obtained from InfoToday. Derwent has a searchable patent glossary and a link to a text patent glossary made by The Minerals, Metals & Materials Society. For principles and sources for patents searching see Free Pint articles by Ron Kamenicki and Stephen Adams.

More recent data can be obtained from the U.S. Patent Office's ftp site.

Updates and changes.