bkelly-labbkelly-labOfficial·8 Agent Skills Included

jkp-data

Global stock characteristics and factor portfolio data pipeline

Generates 406 stock characteristics and factor portfolios from CRSP and Compustat data via WRDS. Automates downloads, industry classification, rolling risk metrics, and value-weighted portfolio construction. Includes pull request review workflows that check code conventions, documentation sync, and test coverage. Eliminates manual data wrangling and slow, error-prone spreadsheet-based financial research.
npx skills add bkelly-lab/jkp-data --all -g -y
Available:

Instructs your AI agent on the data pipeline architecture, Polars coding conventions, testing rules, and step-by-step workflow for making and reviewing changes in this repository.

All Skills in This Repository (8)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install jkp-data?

Run `npx skills add bkelly-lab/jkp-data --all -g -y` in your terminal to install all skills in this suite globally.

What data does jkp-data generate?

It produces global stock returns, 406 firm characteristics, and factor portfolio returns based on the Jensen, Kelly, and Pedersen replication study published in the Journal of Finance.

Do I need a WRDS subscription to use jkp-data?

Yes, the pipeline downloads raw CRSP and Compustat data through your own licensed WRDS account. Pre-computed outputs are also available free at jkpfactors.com.

How do the PR review skills work?

The skills let your agent fetch a pull request, run convention and documentation checks, triage Copilot suggestions, and draft structured review comments or approve and merge.

Can I run the pipeline on an HPC cluster?

Yes. It includes a Slurm batch script and supports persistent WRDS connections to avoid repeated multi-factor authentication prompts on clusters.

Related Repositories in Finance & Accounting

View All in Finance & Accounting