Discounted cash flow (DCF) model for US public companies, built on official SEC EDGAR data.
Status: work in progress. This
mainbranch is the project skeleton. The model is being integrated through a series of pull requests, listed below. Nothing on this branch values a company yet.
- Python package skeleton:
src/dcf, uv project, Python >= 3.11, locked dependencies inuv.lock - MIT license
- GitHub Actions CI: ruff lint, ruff format check and pytest on Python 3.11 and 3.12
- One smoke test that checks that the package imports
- SEC EDGAR client and XBRL extraction of annual 10-K figures
- Core DCF maths: WACC, unlevered free cash flow projection, terminal value, equity bridge, sensitivity grid and reverse DCF
- YAML assumptions and scenarios, market data provider, valuation pipeline and the
python -m dcfcommand-line interface - Microsoft case study notebook and the full README (method, formulas, example output, limitations)
- Finance leases treated as debt-financed capex; latest cover-page share count plus dilution
- Follow-ups from a code audit: corrected warnings, input validation, valuation date and data provenance in the report
This list is a plan, not a feature list. Each item becomes part of main only when its pull
request is merged.
uv sync
uv run ruff check .
uv run ruff format --check .
uv run pytest -qEducational project. Nothing here is investment advice.
MIT, see LICENSE.