Open source cross sectional asset pricing

This repo accompanies our paper: Chen and Zimmermann (2021), "Open source cross-sectional asset pricing"

If you use data or code based on our work, please cite the paper:

@article{ChenZimmermann2021,
  title={Open Source Cross Sectional Asset Pricing},
  author={Chen, Andrew Y. and Tom Zimmermann},
  journal={Critical Finance Review},
  year={2022},
  pages={207-264},
  volume={11},
  number={2}
}

---

Data

If you are mostly interested in working with the data, we provide both stock-level signals (characteristics) and a bunch of different portfolio implementations for direct download at the dedicated data page. Please see the data page for answers to FAQs.

However, this repo may still be useful for understanding the data. For example, if you want to know exactly how we construct BrandInvest (Belo, Lin, and Vitorino 2014), you can just open up BrandInvest.py in the repo's webpage for Signals/pyCode/Predictors/

---

Code

The code is separated into three folders:

1. Signals/pyCode/ Downloads data from WRDS and elsewhere, constructs stock-level signals in Python, and outputs to Signals/pyData/.
2. Portfolios/Code/ Takes in signals and outputs portfolios to Portfolios/Data/. Entirely in R.
3. Shipping/Code/ Used to prepare data for sharing.

We separate the code so you can choose which parts you want to run. If you only want to create signals, you can run the files in Signals/pyCode/ and then do your thing. If you just want to create portfolios, you can skip the signal generation by directly downloading its output via the data page. The whole thing is about 15,000 lines, so you might want to pick your battles.

More details are below

Signals/pyCode Instructions

1. Set up for Creating Signals (Python and R)

• Install Python dependencies:

  cd Signals/pyCode/
  pip install -r requirements.txt

• Install required R packages. [tbc]
• Copy Signals/pyCode/dotenv.template to Signals/pyCode/.env and add your WRDS and FRED credentials.
  • For FRED credentials, request an API key from FRED

2. (Optional) Generate Prep Data

This is only necessary for a handful of signals

If you have bash:

• from Signals/pyCode/
  • run bash prep1_run_on_wrds.sh to copy the prep scripts to the WRDS Cloud
  • wait about 5 hours
    • use qstat to check if it's still running
    • if impatient, check most recent file in ~/temp_prep/log/ on WRDS server.
  • run bash prep2_dl_from_wrds.sh to download the prep data from the WRDS Cloud to Signals/pyData/Prep/

You can alternatively upload to the WRDS Cloud manually, ssh into WRDS, run qsub run_all_prep.sh, and then manually download the prep data.

3. Run the Signals Code

master.py runs the end-to-end Python pipeline. It calls the staged scripts in:

• DataDownloads/ downloads data from WRDS and elsewhere and writes to Signals/pyData/
• SignalMasterTable.py builds the join table used across predictors
• Predictors/ constructs stock-level predictors and outputs to Signals/pyData/Predictors/
• Placebos/ constructs "not predictors" and "indirect evidence" signals and outputs to Signals/pyData/Placebos/

To run:

cd Signals/pyCode/
python master.py

The orchestrator blocks are written to keep running even if a particular download fails (for example due to a missing subscription) so you get as much data as possible. You can track progress in Signals/Logs/.

Portfolios/Code Instructions

master.R runs everything. It:

1. Takes in signal data located in Signals/Data/Predictors/ or Signals/pyData/Predictors/, and Signals/Data/Placebos/ or Signals/pyData/Placebos/
2. Outputs portfolio data to Portfolios/Data/Portfolios/
3. Outputs exhibits found in the paper to Results/

It also uses SignalDoc.csv as a guide for how to run the portfolios.

To run:

• Option 1 - Command line:

  cd Portfolios/Code/
  Rscript master.R

• Option 2 - RStudio: Open master.R in RStudio and click "Source" or press Ctrl+Shift+S (Cmd+Shift+S on Mac)

Before running: You must set pathProject in master.R (line 30) to your project root directory (where SignalDoc.csv is located). If using RStudio, pathProject = paste0(getwd(), '/') should work automatically.

By default the code skips the daily portfolios (skipdaily = T), and takes about 8 hours, assuming you examine all 300 or so signals. However, the baseline portfolios (based on predictability results in the original papers) will be done in just 30 minutes. You can keep an eye on how it's going by checking the csvs outputted to Portfolios/Data/Portfolios/. Every 30 minutes or so the code should output another set of portfolios. Adding the daily portfolios (skipdaily = F) takes an additional 12ish hours.

Minimal Setup

To get started quickly, master.R will create portfolios for Price, Size, and STreversal in Portfolios/Data/Portfolios/. There are a couple ways to set up this signal data:

• Run the code in Signals/pyCode/ (see above).
• Download Firm Level Characteristics/Full Sets/PredictorsIndiv.zip and Firm Level Characteristics/Full Sets/PlacebosIndiv.zip via the data page and unzip to Signals/Data/Predictors/ and Signals/Data/Placebos/.
• Download only some selected csvs via the data page and place in Signals/Data/Predictors/ (e.g. just download BM.csv, AssetGrowth.csv, and EarningsSurprise.csv and put them in Signals/Data/Predictors/).

Shipping/Code Instructions

This code zips up the data, makes some quality checks, and copies files for uploading to Gdrive. You shouldn't need to use this but we keep it with the rest of the code for replicability.