Suppose you have repeated observations for participants in an experiment on the relationship between exercise and weight. What technique would help you address serial correlation in their measurements?
A. Heteroskedastic robust standard errors
B. Probit regressions
C. Clustering the standard error by observation
D. Clustering the standard error by person
Answer: D. Clustering the standard error by person