# RE: st: ci question

 From "Nick Cox" <[email protected]> To <[email protected]> Subject RE: st: ci question Date Fri, 11 Apr 2008 18:25:29 +0100

```See also

SJ-7-1  gr0026  . . . .  Stata tip 42: The overlay problem: Offset for
clarity
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
J. Cui
Q1/07   SJ 7(1):141--142                                 (no
commands)
tip for graphing several quantities on a continuous axis

Chris Witte
Subject: Re: st: ci question

thanks for the example Svend!
I was thinking of doing something like your code,

"replace race=race-0.03 if smoke==0
replace race=race+0.03 if smoke==1"

for offsetting the plots, and this code will take the error out of my
"trial and error" approach to getting the appearance I want.  I'll make
use of your xlabel code as well!

Svend Juul

Chris wrote:

I would like to know how to take the means and standard errors
calculated from the "ci" command, and plot the means with brackets
placed at plus/minus the standard error.

1) So, I first need to know how to get these estimates into a
dataset for use.

2) I found the "serrbar" command, which will do a basic plotting,
but what I'd really like is to have a way to group these estimates,
much like the way you can with "graph box" and the "over" option.
Is this possible?

and later:

... is there a way to horizontally offset the rcap plots within a
single x-axis value so that the rcaps do not overlap?

===================================================================

I find -statsby- most convenient to extract a condensed dataset
with the estimatess of interest. The example also demonstrates
a way to horisontally offset the rcaps:

webuse lbw.dta, clear
statsby mean=r(mean) se=r(se), by(race smoke) clear : ci bwt
generate ub = mean + se
generate lb = mean - se
replace race=race-0.03 if smoke==0
replace race=race+0.03 if smoke==1

twoway (rcap lb ub race)(scatter mean race) , ///
xlabel(1 "white" 2 "black" 3 "other")      ///
ytitle("Birthweight (grams)")              ///
legend(off)

*
*   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/
```