Here's a short tutorial and example of an Event Study, a popular and flexible causal inference model. Event study models can be used for a range of business problems including estimating:

⏺️ Excess stock price returns relative to the market and competitors
⏺️ The impact on KPIs across populations with staggered rollouts 
⏺️ Impact estimates that change over time (e.g. rising then phasing out)

Full video here: https://youtu.be/saSeOeREj5g

In this video, I first describe features of the Event Study, then code an example in python using the yahoo finance API to obtain stock market data. There are many questions you could ask, but in this case, I asked whether JP Morgan had excess market returns from the Nov 5 election results relative to its banking peers. 

At the end of the video, I go into decisions that the Data Scientist must make while modeling, and how the results can (i) change dramatically, and (ii) completely change the interpretation. As with other models, it's really important for that the analyst or data scientist not just blindly use the model but understand how each of their decisions can change results and interpretations. 

If you appreciate this content, the best way to support is some combination of following here, subscribing on youtube, or otherwise engaging with these posts and videos to let the algorithms know these are useful. 🙏 

 #datascientist  #datascience  #causalinference  #eventstudy  #datasciencetutorial