I have to find a "99% confidence bound" for a standard deviation. This is not hard. The only question I have is whether this is finding the $\chi^2_{.99}$ value or just the upper bound for the 99% confidence interval (between $\chi^2_{.005}$ and $\chi^2_{.995}$). Any help would be appreciated. Thanks in advanced!

  • 1
    $\begingroup$ $\chi^2_{0.99}$. Th wording asks for that, and not for a symmetric confidence interval. For the variance in particular, one is seldom interested in a confidence interval in the small tails on both sides sense. $\endgroup$ – André Nicolas Mar 30 '14 at 23:32

The upper confidence bound would be the $\chi^2_{0.99}$ value. As a Google seach will show, the term is in fairly common use, and has a standard meaning: "the" point $a$ such that $F_X(a)=0.99$, the point that has $1\%$ of the area in the right tail.

In particular, the $99\%$ upper confidence bound is not the upper limit of a $99\%$ confidence interval with $0.005$ in each tail.

For variance particularly, upper confidence bounds are the usual quantity of interest. One wants protection against the variance being "too large."


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.