My stats book (Gerstman's Biostatistics for Public Health) has the following formula for the confidence interval of the slope parameter beta. The formula indicates that the denominator for the SE of beta should use $n-2$. However, the example that immediately follows in the book uses $n-1$ in the denominator of the SE of beta. Is this a mistake? Is there missing info? Any thoughts would be much appreciated. Thanks!

enter image description here


Yes, there is a mistake. The CI for $\hat{\beta}$ involves degree of freedom $n-2$, not $n-1$. Note that, the explanation has more typos. In the second bullet-point, it finds $t_{n-2}$, but uses as if it was $t_{n-1}$ in the third bullet point.

  • $\begingroup$ Thank you so much! This is what I thought, but was just concerned there was something I was missing. Again, thank you! $\endgroup$ – Stats TA Nov 8 '19 at 16:13
  • $\begingroup$ Done. Thanks again! $\endgroup$ – Stats TA Nov 9 '19 at 4:00

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.