--- From: "STATA Dndee" <[email protected]>
> > > I want to rank institutions for and outcome given in
> > > outcome% based on Monte Carlo simulation and produce a table
> > > like table 2 in http://plus.maths.org/issue46/risk/index.html.
> > > The method is also described by Spiegelhalter in BMJ, 1998,
> > > 316(1701-1705).
--- On Sun, 18/1/09, Martin Weiss <[email protected]> wrote:
> > Try -help simulate- for a start!
--- STATA Dndee <[email protected]> wrote:
> I tried it before but couldn't understand the commands in Help file.
> I arranged the data with the CIs.
>
> +-------------------------------------------------------+
> | inst cases total outcome lci uci se |
> |-------------------------------------------------------|
> 1. | f 71 273 26.01% 20.8% 31.2% 2.7% |
> 2. | e 259 809 32.01% 28.8% 35.2% 1.6% |
> 3. | c 121 345 35.07% 30.0% 40.1% 2.6% |
> 4. | d 262 726 36.09% 32.6% 39.6% 1.8% |
> 5. | b 192 480 40.00% 35.6% 44.4% 2.2% |
> |
> 6. | a 120 204 58.82% 52.1% 65.6% 3.4% |
> 7. | mean 713 2153 33.12% 31.1% 35.1% 1.0% |
> +-------------------------------------------------------+
I have a very hard time understanding what you want to do. My best
effort on this leads me to believe that you don't want to do a
simulation (or you have already done one) but instead you want to know
how to make a certain graph (which you confusingly called table 2). If
that is the case see the example below:
*------------- begin example ---------------------
clear
input str4 inst cases total outcome lci uci se
f 71 273 26.01 20.8 31.2 2.7
e 259 809 32.01 28.8 35.2 1.6
c 121 345 35.07 30.0 40.1 2.6
d 262 726 36.09 32.6 39.6 1.8
b 192 480 40.00 35.6 44.4 2.2
a 120 204 58.82 52.1 65.6 3.4
mean 713 2153 33.12 31.1 35.1 1.0
end
gen order = -outcome
egen yaxis = axis(order) ///
if inst!="mean", label(inst)
replace yaxis = 0 if inst == "mean"
label define yaxis 0 "mean" , modify
label value yaxis yaxis
twoway rcap lci uci yaxis, ///
horizontal || ///
scatter yaxis outcome, ///
ylab(0/6,valuelabels) ///
ytitle("institution") ///
legend(off)
*--------------- end example -----------------
(For more on how to use examples I sent to the Statalist, see
http://home.fsw.vu.nl/m.buis/stata/exampleFAQ.html )
Hope this helps,
Maarten
-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands
visiting address:
Buitenveldertselaan 3 (Metropolitan), room N515
+31 20 5986715
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
