Example code of simple things one can do with Open Source Asset Pricing data, created with our open source code.
Each file is an independent script that automatically downloads data from the internet. You just need the googledrive R package and a Google Drive account.
Scripts are listed from lightest / fastest (top) to heaviest / slowest (bottom).
| File | Description |
|---|---|
plot_anomaly.R |
Downloads selected long-short returns and plots along with sample and publication dates |
openap_vs_frenchweb.R |
Downloads BM and BMdec portfolios and compares to Ken French's data |
daily_covid_demo.R |
Downloads daily returns (all signals) and plots performance in March 2020 |
FF1993_style_implementation.R |
Downloads selected characteristic, constructs Fama-French 1993 style factor (building on 2x3 sort with Size), compares with Ken French's HML. (2x3 implementations are part of our dataset as of March 2022) |
dl_signal_add_crsp.R |
Creates the data you need for machine learning stuff. Downloads all downloadable predictor characteristics (1.5 gigs zipped), downloads CRSP signals from WRDS, merges, and saves to disk (5.6 gig csv). |
mclean_pontiff_main.R |
Replicates the fact that returns are mostly there out-of-sample (McLean and Pontiff 2016) |
old_vs_new_returns_check.R |
Compares returns from two different releases |
- Assumes working directory is the same folder the file is in.
- Creates folder
temp/in working directory and puts output and intermediate files there - Links to Google Drive URLs may need to be updated
- For the March 2022 Release use https://drive.google.com/drive/folders/1O18scg9iBTiBaDiQFhoGxdn4FdsbMqGo
- For the April 2021 Release use https://drive.google.com/drive/folders/1I6nMmo8k_zGCcp9tUvmMedKTAkb9734R