The spurious regression trap
You regress one random walk on a second, completely independent random walk (they share no common driver at all).
What tends to happen to the regression's -statistic and , and why? How should the analysis be done instead?
Show a hint
Standard regression inference assumes stationary, well-behaved errors. Random walks violate that. Think about what happens to the usual standard-error formula when both series wander without bound.
Your answer
This one is open-ended. Work it through, then check your reasoning against the full solution.