GRIFFIN / ammm 3.3.3.dev0

Marketing mix modelling

Build, assess and interpret Bayesian marketing mix models in Python.

Get startedInstall the library and build your first model. Choose an estimatorUnderstand the time-series, FE and CRE contracts. API referenceFind Python entry points and configuration details.

ammm is a Bayesian MMM library built on PyMC and PyTensor.

Version 3.3.2 is a breaking rename to the ammm distribution and import package. Update existing integrations using the upgrade guide. Load saved models and runs from earlier releases with the release that created them.

The public PanelMMM API includes released named presets for one aggregate time series, fixed effects (FE), and correlated random effects (CRE) panels with one unit dimension, such as geography. Start with Choose an Estimator before preparing a panel model. The random-effects (re) preset remains release-gated.

Documentation Sections

  • Agency Workflow - Minimum evidence requirements, technical review and the review record
  • Methodology - Worked notes on client questions that need designs beyond the default additive MMM
  • Getting Started — Installation, quickstarts, first model
  • Data Preparation — Input data requirements and layout
  • Model Specification — Estimator choice, PanelMMM equation, transforms, priors, and calibration
  • Model Fitting — Fitting, prior predictive checks, save/load
  • Post-Modeling — Diagnostics, contributions, response curves, export
  • Optimization — Budget allocation and interpretation
  • Scenario Planning — Planner specifications, library service and comparison outputs
  • Pipeline Runner — Structured runner, YAML config, staged outputs
  • FAQ — Econometrics explainers for practitioners
  • Contributing — Architecture, development setup, testing
  • API Reference — Module and class reference