That it is a user-written programme is just a fact; mentioning that
signals to others that they might need to install it before they can use
it. In this field most of the good stuff is user-written, but that is
not in itself a problem.
Others know more about this than I do but I imagine that you should use
-glcurve-, as already advised, generate new variables containing curve
coordinates, and then "subtract" one curve from another. I guess that
subtraction might at worst entail interpolation to ensure that curves
were measured for the same coordinates. I can't advise further.
Nick
[email protected]
Shehzad Ali
Thanks, Nick. You rightly pointed out that it's a user written
programme.
The reason I wanted to use -clorenz- is that it plots the difference
between
concentration curves. I don't think -glcurve- does that. Do you know of
any
other way of plotting the difference between concentration curves?
Nick Cox
-clorenz- is a user-written program from SSC. (As I often point out, you
are asked to explain where user-written programs you refer to come
from.)
I've not used it ever, but a glance at the help file indicates that
there is no way that it will support your syntax. The syntax is of
standard form
syntax varlist [, <options> ]
so that I do not know why you expect it to support
syntax (varlist , <options>) (varlist, <options>)
The program also seems to lack some standard features (e.g. support for
-if- or -in-) and to be written in sub-standard style (e.g. uses
variable names that may already be in use).
I suggest that you check out -glcurve- by Philippe van Kerm and Stephen
Jenkins on SSC.
That's well written, well documented and well supported.
Nick
[email protected]
Shehzad Ali
Maybe I am doing something wrong with the syntax here. I am trying to
plot
the difference between concentration curves using - clorenz -. Can I
plot
two differences on the same graph? Here is what I am trying to do but am
getting syntax error.
clorenz (x1 y1, hweight(weight) rank(rank1) dif(c1)) (x2 y2,
hweight(weight)
rank(rank2) dif(c1))
*
* 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/
