Abstract: Randomized Controlled Trial (RCT) has been the gold standard for determining the Average Treatment Effect of a program for marketing, medical, economic, political, and policy applications. Uplift modeling is an additional step to uncover individuals who are most positively influenced by treatment through predictive analytics and machine learning by identifying Heterogeneous Treatment Effects. This approach enables us to identify the likely ""persuadables"" in order to maximize treatment impact through optimal target selection. It has gained significant attention in recent years from multiple fields such as personalized medicine, political election, personalized marketing, and personalized healthcare with growing publications and presentations from industry and academic experts across the world.
This tutorial will include both introductory and advanced topics. I will first introduce the concept of uplift modeling, compare with the traditional response modeling method, and review various uplift predictive analytics approaches. Our discussion will extend from experimental data to observational data, by combining uplift modeling with causal inference. I will also discuss the multiple treatment situation where the best treatment for each individual needs to be determined. Prescriptive analytics from optimization will be utilized to handle the uncertainty of uplift model estimates. Applications and methodologies with examples from multiple industries will be used for illustration.
Bio: Victor S.Y. Lo is a seasoned Big Data, Marketing, Risk, and Finance leader with over 25 years of extensive consulting and corporate experience employing data-driven solutions in a wide variety of business areas, including Customer Relationship Management, Market Research, Advertising Strategy, Risk Management, Financial Econometrics, Insurance Analytics, Product Development, Healthcare Analytics, Operations Management, Transportation, and Human Resources. He is actively engaged with causal inference and is a pioneer of Uplift/True-lift modeling, a key subfield of data science.
Victor has managed teams of quantitative analysts in multiple organizations. He currently leads the AI and Data Science Center of Excellence, Workplace Investing at Fidelity Investments. Previously he managed advanced analytics/data science teams in Personal Investing, Corporate Treasury, Managerial Finance, and Healthcare and Total Well-being at Fidelity Investments. Prior to Fidelity, he was VP and Manager of Modeling and Analysis at FleetBoston Financial (now Bank of America), and Senior Associate at Mercer Management Consulting (now Oliver Wyman).
For academic services, Victor has been a visiting research fellow and corporate executive-in-residence at Bentley University. He has also been serving on the steering committee of the Boston Chapter of the Institute for Operations Research and the Management Sciences (INFORMS) and on the editorial board for two academic journals. He is also an elected board member of the National Institute of Statistical Sciences (NISS). Victor earned a master’s degree in Operational Research and a PhD in Statistics and was a Postdoctoral Fellow in Management Science. He has co-authored a graduate-level econometrics book and published numerous articles in Data Mining, Marketing, Statistics, and Management Science literature, and is completing a graduate-level book on causal inference in business.