chapter eleven

11 The impact of a time-related event

 

This chapter covers

  • Learning to identify when to use Synthetic Controls, Regression Discontinuity Designs or Difference in Differences
  • Understanding their assumptions and limitations
  • Using those methods to estimate causal effects

In Chapter 1, we used an example of evaluating the impact of a newly designed website. We learned that the best way to make such an evaluation was using A/B tests. However, as we already know, it is not always possible to perform A/B tests. Therefore, in the final chapter, we revisit how to assess the decision impact by comparing the situation before and after without relying on A/B tests. That is, we focus on understanding the impact of time-related events.

In this chapter, we’ll explore three techniques for estimating the impact of a decision. Each technique works for a different situation. We’ll stick to the following example as our guide in this chapter. As we explore each technique, we’ll modify this example accordingly to understand when the technique works, what assumptions it has, and how to use it to estimate the ATE.

11.1 Which types of data are we going to use?

11.2 Regression Discontinuity Design

11.2.1 Data simulation

11.2.2 RDD terminology

11.2.3 Assumptions

11.2.4 Impact Estimation

11.2.5 RDD in practice

11.2.6 Exercise

11.3 Synthetic controls

11.3.1 Data simulation

11.3.2 Synthetic Controls terminology

11.3.3 Assumptions

11.3.4 Impact estimation

11.3.5 Synthetic Controls in practice

11.3.6 Selecting training and predicting time periods

11.3.7 Exercise

11.4 Differences in Differences

11.4.1 Data simulation

11.4.2 DiD terminology

11.4.3 Assumptions

11.4.4 Impact estimation

11.4.5 In practice

11.5 Chapter Quiz

11.6 Method comparison

11.7 Summary