Explore the Causal Behavioral Modeling Framework and Discrete Choice Modeling of Consumer Demand in this comprehensive 51-minute talk by Minha Hwang, Principal Architect at Microsoft. Delve into the increasing demand for causal ML models of agent behaviors, enabling unboxing of complex black-box models and proper counterfactual simulations. Learn about discrete choice modeling, its applications in Economics, Marketing Science, and Operation Research, and its relevance to ML/AI researchers. Discover two approaches to modeling consumer heterogeneity, techniques for estimating individual-level models using aggregate sales data, and the connection to reinforcement learning. Gain insights into consumer online search behaviors and neural network representations of discrete choice models. Perfect for those interested in marketing data science, causal inference, and explainable AI.
Causal Behavioral Modeling Framework - Discrete Choice Modeling of Consumer Demand
Overview
Syllabus
– Introduction
– Introduction to discrete choice model
– Where we can use it?
– Key ideas
– Static discrete choice model
– Types of heterogeneity
– Market simulation
– Aggregate sales data
– Consumer sequential search model
– Dynamic discrete choice model
– Neural network representation of discrete model
– QnA
Taught by
Data Science Dojo
