Confidence Intervals
Even for quite large values of n, the actual distribution of the mean is significantly nonnormal. Because of this problem several methods to estimate confidence intervals have been proposed.
Let n1 be the number of successes out of n, the total number of trials, and let
be the proportion of successes. Let zα/2 be the 100 ( 1 − α / 2 )th percentile of the standard normal distribution.
- Wald method
A continuity correction of 0.5/n may be added.
- Agresti-Coull method
Here the estimate of p is modified to
- ArcSine method
- Wilson (score) method
The exact (Clopper-Pearson) method is the most conservative. The Wald method although commonly recommended in the text books is the most biased.
Read more about this topic: Binomial Distribution
Famous quotes containing the words confidence and/or intervals:
“If anybody could overthrow the spoils doctrine and practice, Grant is the man. It has been thought impossible hitherto, but I hope with some confidence that he will win.”
—Rutherford Birchard Hayes (1822–1893)
“It can be demonstrated that the child’s contact with the real world is strengthened by his periodic excursions into fantasy. It becomes easier to tolerate the frustrations of the real world and to accede to the demands of reality if one can restore himself at intervals in a world where the deepest wishes can achieve imaginary gratification.”
—Selma H. Fraiberg (20th century)