Replication Data for the Paper "Is there a secular decline in disruptive patents? Correcting for measurement bias"

Working Paper Title: The Illusive Slump of Disruptive Patents

The repository contains replication data and scripts for the paper: 

Jeffrey T. Macher, Christian Rutzer, Rolf Weder,
Is there a secular decline in disruptive patents? Correcting for measurement bias,
Research Policy,
Volume 53, Issue 5,
2024,
104992,
ISSN 0048-7333,
https://doi.org/10.1016/j.respol.2024.104992

The core component of the repository is the file `replication_file.R` which is an R script to replicate all figures and tables of the paper.

The script relies on multiple data files. The datasets contain CD-values for granted USPTO utility patents for the years 1976-2016. All data is available in two formats, `.fst' (from the R package fst) and `.csv'.

Dataset Descriptions

Main Datasets:

1. `dat_cd5_no_trunc`: This dataset contains the CD5 index of patents based on a method without truncation. 

2. `dat_cd5_trunc_1975`: This dataset contains the CD5 index of patents based on a truncation method as in Park et al. (Nature, 2023).

3. `dat_cd5_no_trunc_app_adj`: This dataset contains the CD5 index of patents based on a method without truncation and including citations to patent applications granted by 2021. 

Additional Datasets:

4. `dat_cd5_trunc_1985`: This dataset contains the CD5 index of patents based on a truncation of all backward citations to patents published before 1985.

5. `dat_cd5_trunc_1995`: This dataset contains the CD5 index of patents based on a truncation of all backward citations to patents published before 1995.

6. `dat_cd5_no_trunc_app`: This dataset contains the CD5 index of patents based on a method without truncation and including citations to patent applications granted by 2021, as well as those not yet granted. 

7. `age_bwc_untrunc`: This dataset contains the age of backward citations using untruncated data.

8. age_bwc_trunc: This dataset contains the age of backward citations using truncated data as in Park et al. (Nature, 2023).

9. `age_bwc_untrunc_app_adj`: This dataset contains the age of backward citations using untruncated data and considering citations of patent applications granted until 2021.

10. `dat_cd10_trunc_1975`: This dataset contains the CD10 index of patents based on a truncation method as in Park et al. (Nature, 2023). 

11. `dat_cd10_no_trunc_app_adj`: This dataset contains the CD10 index of patents based on a method without truncation and including citations to patent applications granted by 2021. 

12. `dat_cd2021_trunc_1975`: This dataset contains the CD index as of 2021 of patents based on a truncation method as in Park et al. (Nature, 2023). 

13. `dat_cd2021_no_trunc_app_adj`: This dataset contains the CD index as of 2021 of patents based on a method without truncation and including citations to patent applications granted by 2021. 

To successfully run the `replication_file.R' script, make sure all data files are in the directory and the `mainDir1' variable at the beginning of the script is set to the correct path to where the data is stored. In addition, set the `mainDir2' variable to the folder where you want to store the figures created by the script. 

If you have any questions, please contact christian.rutzer@unibas.ch