DISCERN Version 2.0: Duke Innovation & SCientific Enterprises Research Network

Together with Sharon Belenzon, Larisa Cioaca, Lia Sheer and John Shin

The DISCERN data set provides a match between patents and scientific articles to Compustat firms. The first version (released in 2020) included more than 4,000 firms over the period 1980-2015, and introduced a major extension and improvement to the historical NBER patent data (Hall, Jaffe, & Trajtenberg, 2001). The data set is unique in that it addresses firms’ name and ownership changes, two major challenges in following patent assignments over time. We are currently extending the match to more recent years, improving the quality of matches and transitioning to open access data sources. Expected to be released by the end of 2023. For the original data, click here.