e****t 发帖数: 766 | 1 About Agresti Coull interval for Binomial Proportion, roughly
let p = ( x + 2 ) / ( n + 4 ),
CI is p +/- z * sqrt(p(1-p)/(n+4))
can i build up the test by
p / sqrt(p(1-p)/(n+4)) ~ Z
so the point estimation i should use is ( x + 2 ) / ( n + 4 ) instead of x /
n ?
(if you know some paper talking about this, it will be great)
Thanks a lot in advance. |
|