Dear Statalisters,


I am using the user-written command eclplot to plot the estimates (and CI) of several regressions (one regression per existing values of varable intyear) for 4 different countries. So I have 4 graphs (defined by variable country in the by option). Here is my code:

sum intyear, meanonly
local lowerbound = 5*floor(`r(min)'/5)
local upperbound = 5*floor(`r(max)'/5) + 5
eclplot (estimate min95 max95 intyear), /*
*/ eplottype(scatter) rplottype(rcap) xlabel(`lowerbound'(5)`upperbound', labsize(vsmall) angle(90)) xtitle("") ylabel(,angle(0)) ytitle("Value of {&beta}{sub:5}") /*
*/ ciopts(blcolor(black)) yline(0, lcolor(black)) estopts(sort) by(country,col(2) ixaxes note("") graphregion(color(white))) subtitle(,fcolor(white) lcolor(white))

Attached is what the code above returns

I was wondering if it were possible to add on each plot a line graph of a third variable (call it fertility) whose values would, of course, only exist for the existing values of intyear of each country. Of course, since fertility does not have the same scale than my estimates, I would like it to have its own yaxis. One yaxis for estimate and one yaxis for fertility, thus.

Thanks a lot for your help!

F