AMMM Documentation (Version 2.5.0)
AMMM Documentation¶
Welcome to AMMM
This library provides a practical framework for building, analysing, and optimising Marketing Mix Models using Bayesian methods with PyMC. Whether you’re new to MMM or an experienced practitioner, AMMM provides tools for rigourous, evidence-led marketing attribution.
Key Features: Bayesian inference • Adstock transformations • Saturation curves • Budget optimisation • Prophet seasonality • Model convergence checks • ELPD-based model selection • Transfer entropy analysis • Stationarity and multicollinearity checks
Documentation Sections¶
New to AMMM? Start here for installation, data preparation, and running your first model.
Perfect for: First-time users, quick setup
Practical step-by-step guides for common MMM tasks like configuring models, interpreting results, and optimising budgets.
Perfect for: Completing specific tasks
Deep dives into MMM methodology, model components, statistical concepts, and the theory behind the framework.
Perfect for: Understanding concepts
Technical details including configuration parameters, output file descriptions, and complete API documentation. API pages are generated at build time by AutoAPI.
Perfect for: Detailed specifications
Data validation tools, model diagnostic checks, and guidance for ensuring robust model performance.
Perfect for: Quality assurance
Solutions for common errors, performance tips, debugging strategies, and frequently asked questions.
Perfect for: Solving problems
MMM Workflow¶
graph LR
A[Prepare Data] --> B[Configure Model]
B --> C[Fit Model]
C --> D[Validate Results]
D --> E[Optimise Budget]
D --> F{Satisfied?}
F -->|No| B
F -->|Yes| G[Deploy Insights]
style A fill:#bbdefb,stroke:#1976d2,stroke-width:2px
style C fill:#ffe0b2,stroke:#f57c00,stroke-width:2px
style D fill:#b2dfdb,stroke:#00897b,stroke-width:2px
style E fill:#c8e6c9,stroke:#388e3c,stroke-width:2px
style G fill:#c8e6c9,stroke:#388e3c,stroke-width:2px
Documentation Structure¶
Getting Started: Installation and first model tutorials
How-to Guides: Task-oriented guides for specific objectives
Explanation: Conceptual background and methodology
Reference: Technical specifications and API docs
Diagnostics: Validation and quality assurance
Troubleshooting: Problem-solving and debugging
About this documentation¶
Practitioners: Start with Getting Started → Quickstart, then run the pipeline via
runme.py.Engineers: See Guides and the API Reference. API pages are produced during the build by AutoAPI.
Analysts: Refer to the Output Schema and Guides on interpreting results to work with CSV outputs.