Causal Inference in Marketing: A Practical Toolkit for Panel Data

Two-Volume Set
1. Auflage 2026
ISBN: 978-1-041-40388-3
Verlag: Taylor & Francis

Buch, Englisch, 1072 Seiten, Format (B × H): 178 mm x 254 mm

Causal Inference in Marketing: A Practical Toolkit for Panel Data is a two-volume guide to turning messy marketing panels into credible causal evidence. Written for data scientists, marketing analysts, econometricians, and applied researchers, it connects modern causal inference with the operational realities of advertising, pricing, loyalty, platforms, and marketing effectiveness. The set is distinctive in its design-first treatment of panel data: it begins with estimands, assignment mechanisms, support, and diagnostics before moving to estimators. Readers should be comfortable with regression and applied statistics, but the exposition is built to make the assumptions behind causal claims explicit rather than hidden inside software defaults.

Across the two volumes, readers learn how to choose, implement, diagnose, and report causal designs for marketing measurement. Volume 1 develops the foundations, including potential outcomes, design-based thinking for panels, difference-in-differences, event-study designs, synthetic control, hybrid synthetic control methods, interactive fixed effects, matrix completion, dynamic treatment effects, heterogeneity, interference, and spillovers. Volume 2 extends the toolkit to machine learning for nuisance adjustment and treatment-effect heterogeneity, high-dimensional controls, regularisation, continuous and nonlinear panel models, threats to validity, inference and uncertainty quantification, design diagnostics, and applied marketing workflows. The result is a practical framework for evaluating incrementality, media mix models, geo-experiments, platform data, pricing, promotions, customer lifetime value, retention, and reproducibility while keeping the causal target, data limitations, and business decision aligned.

Zielgruppe

Professional Practice & Development, Professional Reference, and Professional Training

Autoren/Hrsg.

Shaw, Charles

Fachgebiete

• Mathematik | Informatik Mathematik Stochastik
• Wirtschaftswissenschaften Betriebswirtschaft Wirtschaftsmathematik und -statistik

Inhaltsverzeichnis

Part 1: Foundations 1. Why Marketing Panel Data Need Causal Design 2. Causal Frameworks and Panel Notation 3. Design-Based Thinking for Panels Part 2: Differences-in-Differences and Event Studies 4. Difference-in-Differences: From Canonical to Staggered 5. Event-Study Designs Part 3: Synthetic Controls and Hybrid Methods 6. Synthetic Control 7. Hybrid Synthetic Control Methods Part 4: Factor Models and Matrix Methods 8. Interactive Fixed Effects and Matrix Completion 9. Advanced Matrix Methods for Causal Inference  Part 5: Dynamics, Heterogeneity, and Spillovers 10. Dynamic Treatment Effects 11. Interference and Spillovers Part 6: Machine Learning and High-Dimensional Methods 12. Machine Learning for Nuisance and Heterogeneity 13. High-Dimensional Controls and Regularisation 14. Continuous and Nonlinear Panel Models Part 7: Validity, Inference, and Diagnostics 15. Threats to Validity in Marketing Panels 16. Inference and Uncertainty Quantification 17. Design and Diagnostics Part 8: Applications and Future Directions 18. Applications in Marketing 19. Measurement, Platform Data, and Reproducibility 20. Outlook and Open Problems Part 9: Appendices A. Time Series: Recap of Basic Principles B. Stationarity and Cointegration in Panels

Über Autor(innen)

Charles Shaw is a Data Science Director at WPP Media, where he leads econometric measurement and optimisation for global brands. His work focuses on causal inference, econometric measurement, Bayesian modelling, machine learning, and marketing effectiveness. He develops applied frameworks for privacy-constrained attribution, media incrementality, platform effects, dynamic pricing, and scalable causal workflows in commercial settings.