Step 04
Running the analysis
The empirical backstage: specifications and robustness check.
This is a good place to include all the robustness checks I made and explain a bit more about how I actually ran the regressions. But because that tends to be quite lengthy, and since I have a hunch that whoever is interested in this would be rather academic, I'll just refer you to the corresponding chapter in the manuscript.
All in all, running a regression seems straightforward when it's done, but really there are so many micro-choices that influence results and interpretation. Here is an overview of the workflow for the regression, to give you an idea of the choices I had to make and which now seem as given.