[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: confidence intervals around point estimates plotted against time ascategorical variable
Many thanks to Nick Cox about the earlier response. I have another
The dataset contains:
1000 patients (observations), each with the following variables:
date: day and month patient enrolled
x: "1" for using drug x and "2" for not using drug x.
As suggested by NC, I used: egen medx = mean(100*(2-x)), by(month)
I was able to nicely plot medx vs. month, showing what percentage of
patients are using drug x in that given month.
Now, I wish to add 95% confidence intervals around the percentages for
each given month. I've read and downloaded Roger Newson's program and
tried to use rcap, but cannot seem to figure out how to do this easily,
much less do it at all. I've been working on this for over 2 hours, so
thought would ask for help.
Also, when I use box plot, I get data markers on "1" and "0" for a given
Thank you in advance,
* For searches and help try: