Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
"Wood, Mollie Elizabeth" <MollieElizabeth.Wood@umassmed.edu> |

To |
"statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |

Subject |
st: manually calculating confidence interval for a proportion |

Date |
Fri, 12 Jul 2013 21:49:39 +0000 |

I am putting together a teaching example for a beginning programming lab, in which I ask students to manually calculate the upper and lower confidence limits for a variety of point estimates, and compare their results to the stata commands (ci, cii). However, I haven't had much look making it work myself! > I used this code to obtain 90% CIs for a mean: scalar calculated_90ub = r(mean) + invttail(_N - 1,0.05)*LOS_se scalar calculated_90lb = r(mean) - invttail(_N - 1,0.05)*LOS_se dis calculated_90ub dis calculated_90lb > Which worked well, and matched the output of Stata's command, ci [var], level(90) > However, when I tried the below command for the 95% CI around a proportion, the results are quite different the stata command, cii N K, level(95) > scalar calculated_95ub_bin= 157/484 + invbinomial(484,157,.05)*(sqrt((1/484)*(157/484)*(1-(157/484)))) scalar calculated_95lb_bin= 157/484 - invbinomial(484,157,.05)*(sqrt((1/484)*(157/484)*(1-(157/484)))) > dis calculated_95ub_bin dis calculated_95lb_bin > I did experiment with several other related functions (invbinomialtail, etc), none of which accurately reproduced the output from cii. > If anyone has a suggestion as to how I'm going wrong, I'd appreciate it very much! * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/faqs/resources/statalist-faq/ * http://www.ats.ucla.edu/stat/stata/

**Follow-Ups**:**st: No cii error: was "manually calculating confidence interval for a proportion"***From:*Steve Samuels <sjsamuels@gmail.com>

**Re: st: manually calculating confidence interval for a proportion***From:*Steve Samuels <sjsamuels@gmail.com>

**Re: st: manually calculating confidence interval for a proportion***From:*Steve Samuels <sjsamuels@gmail.com>

**Re: st: manually calculating confidence interval for a proportion***From:*Steve Samuels <sjsamuels@gmail.com>

- Prev by Date:
**st: How to refer to results from three mlogit models in a single predictnl command** - Next by Date:
**Re: st: manually calculating confidence interval for a proportion** - Previous by thread:
**st: How to refer to results from three mlogit models in a single predictnl command** - Next by thread:
**Re: st: manually calculating confidence interval for a proportion** - Index(es):