09-14, 12:00–12:30 (Europe/Amsterdam), Qux
The Difference-in-Differences (DiD) methodology is a popular causal inference method utilized by leading tech firms such as Microsoft Research, LinkedIn, Meta, and Uber. Yet recent studies suggest that traditional DiD methods may have significant limitations when treatment timings differ. An effective alternative is the implementation of the staggered DiD design. We exemplify this by investigating an interesting question in the music industry: Does featuring a song in TV shows influence its popularity, and are there specific factors that could moderate this impact?
Difference-in-differences (DiD) is a causal inference method frequently used in empirical research in industry and academia. However, standard DiD has limitations when interventions occur at different times or affect varying groups. This talk will highlight the application of the Staggered DiD method, a more nuanced approach that addresses these limitations, in the context of the music industry. We will try to answer the question of how music features in TV shows affect music popularity and how this effect might change for different types of music using the staggered DiD method. Attendees will gain an understanding of causal inference through observational studies and specifically how the new DiD methods are used through an interesting and original case study.
The talk will be structured as follows:
- Intro to the case (e.g., background on music features on TV, dataset)
- Explanation of the DiD approach and its limitations.
- Introduction to the Staggered DiD method.
- Application of staggered DiD for the case study from the music industry
- Conclusions
- Q&A
Target Audience: The talk would be beneficial for data scientists, researchers, and practitioners interested in causal inference, marketing analytics, and quasi-experimental design. Attendees should have a basic understanding of statistical methods used in data science.
Key Takeaways:
- Understanding of the DiD approach and its limitations in the context of analyses with observational data.
- Insights into the Staggered DiD method and its application.
- Practical knowledge about executing and evaluating DiD studies effectively.
No previous knowledge expected
I am a quantitative researcher and data scientist with a strong background in marketing, economics, and econometrics. My focus is on using data-driven approaches to tackle complex business challenges, uncover valuable insights, and drive impactful decisions. As a Ph.D. candidate in quantitative marketing, I specialize in causal inference, machine learning, and experimental design to address cutting-edge research questions.