This course provides an introduction to causal machine learning for the assessment of the causal effect of some action or intervention, like offering or not offering a discount to customers, on an outcome of interest, like customers' buying decision. The assessment of a causal effect requires that customer groups receiving and not receiving a discount are comparable in background characteristics that also affect their buying behavior (e.g. previous buying behavior, income, education etc.). Machine learning can be used to find such comparable groups by means of statistical models for how the characteristics affect the intervention and the outcome. Such approaches also permit detecting customer groups for whom the effect of the intervention is particularly large as a function of their characteristics. This course first discusses the assumptions underlying causal analysis as well as the usefulness of machine learning for assessing causal effects and considers various algorithms. Using the statistical software "R" and its interface "R Studio", the methods are applied to various real-world data sets.
Objectives
- To understand the idea and goals of machine learning for causal analysis.
- To understand the intuition, advantages, and disadvantages of alternative methods.
- To be able to apply causal machine learning to real world data using the software "R" and its interface "R Studio".
Content
- Basic concepts of machine learning for causal analysis and differences to predictive machine learning.
- Causal analysis based on penalized regression (lasso and ridge regression) and tree-based approaches (causal trees and causal forests).
- Assessing effect heterogeneity based on machine learning to find groups for whom the intervention is most effective.
- Application of all methods to real world data using the statistical software "R" and its interface "R Studio"; course participants may bring along their own data sets.
Date: July 17, 2021
Plan:
Lecture 1: 08:30-10:00
Break 10:00-10:30
Lecture 2: 10:30-12:00
Lunch break 12:00-13:30
Lecture 3: 13:30-15:00
Break 15:00-15:30
Lecture 4: 15:30-17:00