--- Carlo Lazzaro <[email protected]> wrote:
> I was dealing with model aimed at comparing two healthcare programmes
> (A and B; 255 observations each) in terms of their effectiveness.
> I cannot rely on a dataset, but only on a point estimates for A and B
> effectiveness, coming from a clinical trial (programme A=88,89%;
> programme B= 79,76%; difference=9,13%).
The direct answer would be -prtesti 255 .8889 255 .7976-.
Since you were looking for a bootstrap/simulation kind of approach (and
I like that kind of approach since it is often more intuitive) I have
also wrote a simulation. I am assuming that the proportion is referring
to the proportion of people that have not died, or suffered something
else unpleasant, after having gone through either program A or B, i.e.
the underlying data, that you don't have, are 255 cases for each
program where each case could have either a 0 or a 1. So I randomly
created this data, assuming that the found proportions where the true
population proportions. This generated data would be a random sample
from this `population'. I did so many times (2000) and computed the
difference in `sample' proportions each time, thus creating the
sampling distribution of the difference.
*---------- begin example -----------
capture program drop sim
program define sim, rclass
drop _all
set obs 255
gen byte successA = uniform() < .8889
gen byte successB = uniform() < .7976
sum successA, meanonly
scalar A = r(mean)
sum successB, meanonly
return scalar diff = A - r(mean)
end
simulate sim diff=r(diff), reps(2000)
sum diff
centile diff, centile(2.5 97.5)
*---------- 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 Z434
+31 20 5986715
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
___________________________________________________________
To help you stay safe and secure online, we've developed the all new Yahoo! Security Centre. http://uk.security.yahoo.com
*
* 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/