Your problem really is with -concord-, not with -graph-.
On this occasion, the program insists on having the last word
on graph options and ignores -legend(off)-. I can't remember if
there was some subtle reason for that, but it does at this
instant look like a misfeature. I think Tom Steichen wrote
this bit of the code and he may be able to add more.
>>> Kevan Polkinghorne wrote
I have been producing Bland and Altman graphs using the nice concord
program written by Thomas J. Steichen and Nicholas J. Cox. My problem
relates not really to the program but the graph produced. The default is
that the legend is on for the Bland and Altman graph. I would like to
turn it off but for some reason stata it not responding to the graph
option legend(off). I do not get error messages. I have tried moving the
commond option around, either makes no difference or I get a error
message as it is in wrong place. Makes no difference with version 8 or
9. Please see my syntax below. I'm not sure if it is my syntax or
something else. Couldn't find a posting in the archives that describes a
similar problem.
concord loguacrhplc loguacr if cohorthplc==1,
loa(msymbol(circle_hollow) msize(vsmall) mcolor(navy) regline
ytitle("Difference of Log Urine ACR HPLC and" "Immunonephrometry
(mg/mmol)", size(medium) color(black) orientation(vertical)
justification(center) alignment(bottom)) xtitle(Mean of Log Urine ACR
HPLC and Immunonephrometry (mg/mmol), size(medium) color(black)
orientation(horizontal) justification(center) alignment(middle))
legend(off) scheme(s1manual) graphregion(fcolor(white) lcolor(none)
ifcolor(white) ilcolor(none)) plotregion(fcolor(white) lcolor(none)
ifcolor(white) ilcolor(none)))
*
* 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/
