Common methods to calculate confidence band for binomial distribution
Peer uest
Cancel
Forget passwd??
blog
4
Common methods to calculate confidence band for binomial distribution

The textbook confidence interval for binomial trial is given by

$CI=\hat{p}\pm z\sqrt{\frac{\hat{p}(1-\hat{p})}{\tilde{n}}}$
It is widely recognizedthat the actual coverage probability of the standard interval is poor for p near 0 or 1.Below are some alternatives that give a much better confidence bands

Agrestii-Coull Interval

The confidence band is given by

$CI_{AC}=\tilde{p}\pm z\sqrt{\frac{\tilde{p}(1-\tilde{p})}{\tilde{n}}}$

where

$\tilde{n}=n+z^2$

and

$\tilde{p}=\frac{1}{\tilde{n}}(x+\frac{z^2}{2})$

Wilson Interval

$CI_W =\frac{x+z^2/2}{n+z^2}\pm \frac{z\sqrt{n}}{n+z^2}\sqrt{\hat{p}(1-\hat{p})+z^2/(4n)}$

Exact Confidence Band

For small sample size, exact confidence band should be used.

×
×
×
×