Confidence interval- Please help.

[Sydney Lungu]

Dear stata friends,

My boss has asked to calculate the confindence interval because I produced an output like the one below. How do i do this? What do i need to do this? please help


.tab AE50

AE50 | Freq. Percent Cum.
------------+-----------------------------------------
No | 1,139 70.40 70.40
Yes | 479 29.60 100.00
------------+-----------------------------------------
Total | 1,618 100.00

Please help help

Sydney

[Joseph Coveney]

My boss has asked to calculate the confindence interval . . . How do i do this?

cii proportions 1618 479

More generally:

help ci

[Sydney Lungu]

Thanks so much, this helped and it bworked

[Sydney Lungu]

Just a follow questions what is the difference when one codes;

a. cii proportions 1618 479
b. proportion AE50
c. ci means AE50

[Dirk Enzmann]

There are different methods for calculating a confidence interval for a binomial proprortion -- by default (a) calculates the Clopper-Pearson "exact" interval (not that this is better because of the term "exact"), (b) calculates the intervals based on logits, and (c) calculates the "normal" or so-called "Wald" interval.

Their confidence intervals have different coverage properties, I recommend to have a look at the options for each command. More detailed information can be found in: Brown, L. D., Cai, T. T., & DasGupta, A. (2001). Interval estimation for a binomial proportion. Statistical Science, 16(2), 101-133 (they would recommend the Wilson or the Jeffreys interval).

[Dirk Enzmann]

Here is a direct link to the Brown et al. article: https://projecteuclid.org/journalArt...s%2F1009213286

[Leonardo Guizzetti]

Newcombe has a very good article about interval estimation for single proportions and differences between two independent proportions that are worthy reads. The Score (Wilson) interval has good properties, but there are some areas where there are different norms, such as is clinical trials where the binomial (exact) CI is preferred (if not conservative).

Newcombe, R.G. (1998), Two-sided confidence intervals for the single proportion: comparison of seven methods. Statist. Med., 17: 857-872. https://doi.org/10.1002/(SICI)1097-0...777>3.0.CO;2-E

Newcombe, R.G. (1998), Interval estimation for the difference between independent proportions: comparison of eleven methods. Statist. Med., 17: 873-890. https://doi.org/10.1002/(SICI)1097-0...779>3.0.CO;2-I