I have been using the Royston and Cox -mrunning, Stata 9.1, windows xp, a
very helpful addition to Stata.
I made a graph using this code:
use cl_anemia_ra1, clear
set seed 239
getran whoanem sex age
mrunning whoanem age if ruse,adj(sex) ci yline(.3152 ) nopt ylabel(0[.1].7)
///
xtitle("Age (years)") ///
ylab(0 "0%" .1 "10%" .2 "20%" .3 "30%" .4 "40%" .5 "50%" .6 "60%" .7 "70%"
.8 "80%" .9 "90%" 1 "100%", ///
angle(0)) title(The effect of age on anemia prevalence in RA) ///
scheme(s2mono) clwidth(medthick) clpattern(solid dot dot) ///
combine(saving(ageprevanem,replace))
graph export ageprevanem.wmf,replace
When I combined this graph with another using the following code:
graph combine age.gph ageprevanem.gph, altshrink
I noticed the graph from -mrunning was physically smaller, such that the
two graphs from the immediately above command were not of equal size. I
combined other graphs and was able to determine the problem was with
-mrunning-. I wonder if Nick Cox might shed some light on this?
Thanks,
Fred
Fred Wolfe
National Data Bank for Rheumatic Diseases
Wichita, Kansas
Tel (316) 263-2125 Fax (316) 263-0761
fwolfe@arthritis-research.org
