I am looking for ideas for research projects to give my students in a module on causality and DAGs (third-year undergraduate and master level).
I had a couple of ideas. For example, one could, for example, provide a dataset and ask students to suggest a DAG that would be a likely candidate to explain it. Or, one could ask the students to start from a DAG and generate the data that would reflect the model. And then from the data try to "reverse engineer" the model.
Any specific ideas would be most appreciated! Thank you :-)
For example, consider the following DAG, where the outcome variable is whether or not an Uber passenger accepts an offer for a journey - the outcome variable depends on the price of the journey and the expected waiting time. Students were requested to use this model (or develop something of their own), generate appropriate data and explore the data they generated to "discover" the model. What do you think? Any ideas along this line?