PhD Data Science Intern – Media Mix Modeling

This is a fully remote position, open to applicants in United States.

📋 Description

• Conduct research and assess statistical and econometric methods for Media Mix Modeling (MMM) and measuring marketing effectiveness.

• Create, test, and refine MMM algorithms throughout the entire modeling lifecycle.

• Utilize time-series, panel, and observational marketing datasets to construct reliable models of media response and business outcomes.

• Investigate media response curves, saturation effects, adstock and carryover effects, incrementality and causal inference, channel interactions and synergies, seasonality, trends, external factors, regularization, variable selection, uncertainty estimation, statistical inference, as well as Bayesian and frequentist modeling techniques.

• Establish model diagnostics and validation frameworks to evaluate model stability, predictive performance, statistical significance, and business interpretability.

• Perform simulations and experiments to analyze algorithm behavior under varying data-generating conditions.

• Compare various modeling techniques and pinpoint opportunities to enhance model accuracy, robustness, and interpretability.

• Convert research insights into production-ready algorithms and analytical workflows.

• Engage with actual client datasets and tackle the practical challenges of implementing MMM with imperfect business data.

• Collaborate with senior data scientists to document methodologies, assumptions, limitations, and results.

• Contribute to the advancement of next-generation MMM capabilities at FocusKPI.

⛳️ Requirements

• PhD in Statistics, Economics, Econometrics, Applied Mathematics, Data Science, or a closely related quantitative discipline.

• Strong theoretical background in statistical modeling, econometrics, regression and multivariate analysis, time-series analysis, probability and statistical inference, and optimization.

• Comprehensive understanding of causal inference and observational data.

• Proven ability to develop statistical models from start to finish, including problem formulation, data preparation and feature engineering, model specification, estimation, diagnostics, validation, interpretation, and implementation.

• Proficient programming skills in Python.

• Experience working with large and intricate datasets.

• Capability to translate mathematical and statistical ideas into practical algorithms.

• Strong analytical and problem-solving capabilities.

• Ability to work autonomously while maintaining close collaboration with senior technical team members.

• Direct experience with Media Mix Modeling (MMM) is preferred.

• Background in marketing measurement, marketing analytics, or advertising data is preferred.

• Familiarity with Bayesian hierarchical models is preferred.

• Experience with causal inference, experimentation, or uplift modeling is preferred.

• Knowledge of time-series econometrics is preferred.

• Familiarity with Bayesian inference / MCMC, state-space models, regularization, constrained optimization, nonlinear regression, response curve estimation, and Monte Carlo simulation is preferred.

• Experience with modern statistical computing frameworks such as PyMC, Stan, NumPyro, JAX, scikit-learn, statsmodels, or equivalent is preferred.

• Proven ability to take research concepts and transform them into reusable production code is preferred.

🏝️ Benefits

• Full-time, paid internship.

• Opportunity for internship-to-full-time transition.

• Flexible remote work arrangement.