Skip to main content
Ctrl+K

macroforecast

  • Getting started
  • User guide
  • Reference
  • Replications
  • Help
  • GitHub
  • PyPI
  • Getting started
  • User guide
  • Reference
  • Replications
  • Help
  • GitHub
  • PyPI

Section Navigation

1. Using the package

  • Simple API
    • Simple API quickstart
    • Run an experiment
    • Compare models and parameters
    • Read results
    • Use FRED-SD
  • Simple API quickstart
  • Run an experiment
  • Compare models and parameters
  • Read results
  • Use FRED-SD
  • Recipe API
  • Recipe Gallery
  • Recipe Layer Contract
  • Data Layer
  • Data Policies
  • Recipe Defaults
  • Runtime Support Matrix
  • Understanding Recipe Output
  • FRED Datasets in Recipes

2. Bring your own

  • Bring Your Own Data
  • Custom Function Quickstart
  • Custom Extensions
  • Target Transformer Contract
  • Partial Layer Execution
  • User guide
  • Recipe Gallery

Recipe Gallery#

The runnable gallery lives in examples/recipes/. Gallery recipes must use current top-level layer keys, not the legacy path: 1_data_task wrapper.

Use:

from pathlib import Path
import macroforecast as mf

result = mf.run(Path("examples/recipes/l4_minimal_ridge.yaml"))

or:

macroforecast run examples/recipes/l4_minimal_ridge.yaml -o out/l4_minimal_ridge

Recommended first recipes:

Recipe

Use it for

l4_minimal_ridge.yaml

Small end-to-end custom-panel recipe.

l6_standard.yaml

AR-p benchmark versus ridge with DM-HLN and MCS blocks.

l6_full_replication.yaml

Broader L6 statistical-test coverage.

goulet_coulombe_2021_replication.yaml

Paper-style replication scaffold.

Do not describe optional extras unless the recipe actually imports that family. Keep this page generated or smoke-test-backed.

previous

Recipe API

next

Recipe Layer Contract

Edit on GitHub
Show Source

© Copyright 2026, NanyeonK.

Created using Sphinx 9.0.4.