Paired versus two-sample t-test, when to pair
Asked at Point72
You want to test whether two conditions differ in mean: for example, the same set of stocks measured before and after a signal, versus two separate groups of stocks.
When do you use a paired t-test versus an unpaired (two-sample) t-test, and why does pairing help?
Show a hint
If each observation in group A has a natural partner in group B, you can test the differences directly. What does that do to shared variability?
Your answer
This one is open-ended. Work it through, then check your reasoning against the full solution.