Dale Steele
> I have been struggling to "overlie" two sets of data on the same
> scatterplot matrix. I have two stata data files. The
> first contains the
> raw data from an experiment. The second file contains
> calculated predicted
> values which I would like to plot as a line on top of the
> observed values in
> the first file.
>
> Various permutations of the -merge- command have failed
> to merge the two files.
>
> I've made plots of each using:
> graph response resistnc, by(idnum) total jitter(2)
> ylabel(0, 2 to 10)
> and
> graph R resistnc, by(idnum) connect(l) ylabel(0, 2 to 10) total
>
> Suggestions?
>
> Details follow..
>
> Each subject given 6 stimuli, response to each stimulus
> measured five
> times. Thus the file looks like:
>
> list idnum resistnc response
>
> idnum resistnc response
> 1. 401 0 0
> 2. 401 0 0
> 3. 401 0 0
> 4. 401 0 .7
> 5. 401 0 0
> 6. 401 2.5 1.9
> 7. 401 2.5 1.5
> ...
>
> The second file contains a predicted response (R) based on
> a predicted curve
> for each subject.
>
> . list idnum resistnc R
>
> idnum resistnc R
> 1. 401 0 .14
> 2. 401 2.5 2.489528
> 3. 401 5 3.538459
> 4. 401 10 5.055678
> 5. 401 20 7.25025
> 6. 401 30 8.963791
> 7. 402 0 .4
> 8. 402 2.5 1.289712
> 9. 402 5 2.060106
> 10. 402 10 3.497576
> 11. 402 20 6.179739
>
Note that Dale wants a matrix of scatterplots, but not one in
the sense of -graph, matrix-. The files having been merged,
this would seem to be something like
. sort idnum
. graph response R resistnc, by(idnum) total jitter(2)
ylabel(0(2)10) c(.l) sy(oi)
The key question thus appears to be what Dale tried with -merge-
and why it didn't work. The only clue given here is that
-resistnc- appears in different form for -idnum- 401.
Nick
[email protected]
*
* 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/