Overcoming selection bias with econometrics
Here’s something we haven’t explored much on Enterprise: Econometrics, or the use of statistics to make sense of economic data, has an uncanny power to separate fact from fiction. The use of regression models is often able to overcome the notorious “selection bias,” which makes the “search for causal knowledge” all the more complicated. In this episode of YouTube series "Mastering Econometrics" (watch, runtime: 9:32), MIT professor Josh Angrist breaks down how the discipline can overcome selection bias. Most notably, regression allows econometricians to look at the effect of one variable on another while isolating all others, creating a situation known as ceteris paribus, which is Latin for "other things equal.” Angrist illustrates this using the example of wage differences between graduates of state-owned universities and those who studied at private universities.