Code & Data

Patent data

• KPSS patent data updated to 2024
• If you want to replicate the results of KPSS (2017), see this replication kit
• Mike Woeppel has some additional cleaned data from USPTO

Code

• Here’s a little tool I built for mapping SIC codes to Fama-French industry classifications in Python.
• Long ago, I posted some SAS code to estimate regressions with Fama-Macbeth, clustered standard errors, or fixed effects in SAS. This code complements the Stata code available on Mitchell Petersen’s Programming Advice page. I haven’t used SAS in many years and cannot say whether this is still the best way to do things.
• SAS code to estimate PIN (as in Easley, Kiefer, O’Hara & Paperman, 1996, Journal of Finance). This is really old!

GitHub projects

Occasionally I’ll build a little tool to do something I need. Here are two:

• ff_industries, a small python tool to grab SIC-industry mappings for the Fama-French industries
• OLS_VIZ, a website for illustrating key information about regression. I built this by telling Codex what I wanted, and poof! It’s about 1400+ lines of Javascript, which I don’t even know.