st: RE: a "byable" cii

 From "Sayer, Bryan" To "'statalist@hsphsun2.harvard.edu'" Subject st: RE: a "byable" cii Date Thu, 6 Feb 2003 15:29:48 -0500

```Gee, personally I would suggest just doing them in the spreadsheet, if that
is the way the data came.  (Standard error of proportion = sqrt[p*(1-p) / n]
, 95% c.i. = p +/- 1.96*se(p))

Bryan Sayer
Statistician, SSS Inc.
bsayer@s-3.com

-----Original Message-----
From: VISINTAINER PAUL [mailto:VISINT@NYMC.EDU]
Sent: Thursday, February 06, 2003 3:05 PM
To: 'statalist@hsphsun2.harvard.edu'
Subject: st: a "byable" cii

I was recently asked to compute confidence intervals for several proportions
(more than 100).  The data came in a spreadsheet in two columns: numerator
(num) and denominator (den).  I thought I could quickly run them using
either ci or cii.  However, it appears that ci doesn't read summary data and
cii is not "byable".  Despite my best efforts, I couldn't devise a program
that would read more than the first line of the data.

My very inelegant solution to the problem was this:  I converted the two
columns of data into a single row of data with 200 columns:  (e.g., num1
num2 . . . num100 den1 den2 . . . den100).

Then I ran the following program:

forvalues x=1/100 {
cii den`x' num`x'
}

While it worked, it was really cumbersome converting the data.  I suspect
(hope) there is an easier way to do this.  Any suggestions?

-p
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
```