Toy projects and useful resources extracted from other projects:

  • Stata Code Viewer (code dependencies)
  • Historical German Patent Technology Classes
  • Web of Science subject area - technology class correspondences

Stata Code Viewer

Stata Code Viewer Flowchart

Visualize flows of Stata code. This helps to create a visual overview over large Stata projects. This program, written in Python, parses the most frequent input/output commands in Stata code and automatically visualizes the flow of the a project. With this, it is possible to keep track of your own project, detect errors and execution loops as well as explore the code by others (Coauthors ;-)) more easily.

In a nutshell, visualize which .do file:

  • Uses which data source
  • Creates which files

Link to the repository

Historical German Patent Technology Classes

Stata Code Viewer Flowchart

This is a digitized list of description of patent technology classes and sub-classes. It is intended to give researchers working with historical German patent data easier access to essential information not otherwise available on the web.

Link to the repository

(German) Dies ist ein Daten-Repository, dessen Ziel es ist, Forschern eine digitale Version der Beschreibungen historischer deutscher Patentklassen, einschließlich der Gruppeneinteilung der Patentklassen, zur Verfügung zu stellen.

Web of Science subject area - technology class correspondences

Correspondence between Web of Science subject areas and various patent technology classes, based on scientific non-patent literature citations in patents. This is a probabilistic crosswalk, based on empirical data.

Currently, CPC class and subclass as well as area34 and their correspondence to Web of Science subject codes as well as OECD fields of science are considered. If extensions are required, please add an issue in the repository.

The original data construction is described in our Science Advances article and the corresponding supplement. Please cite that article when using data from this repository. Generally speaking, a high-quality match between non-patent literature references in USPTO and EPO as well as WIPO patents was executed. Here patents in the 1980-2012 time range are considered. All data is aggregated at the level of patent families and only patent families with at least one granted member at the USPTO or EPO are considered.

Link to the repository