How to move estimates to center of the graph?

[Matthew Williams]

Hi all,

I am trying to plot estimates and their associated 95% CI into one graph. Actually, I successfully made a graph, but the distance of estimates between country A and B is huge. So, I want to narrow down that distance by moving the estimates of the two countries to the graph center, any advice is highly appreciated. Thank you.
Data

clear
input float(id country beta ub lb)
 1 0  -.085 -.068 -.101
11 1 -.115  -.089 -.141
end
label values country country
label def country 0 "A", modify
label def country 1 "B", modify

Code

twoway    (scatter beta country, msymbol(s))                    ///
        (rspike ub lb country), xla(0/1, valuelabel noticks)  ///
        legend(label(1 "Estimates") label(2 "95% CI"))

[Andrew Musau]

Just extend the x scale range. At the moment you have the values 0 and 1.

xsc(r(-2 4))

[Øyvind Snilsberg]

xscale,

twoway  scatter beta country, msymbol(s) || ///
        (rspike ub lb country), xla(0/1, valuelabel noticks)  ///
        legend(label(1 "Estimates") label(2 "95% CI")) ///
        xscale(r(-3 4))

xscale and xsize,

twoway  scatter beta country, msymbol(s) || ///
        (rspike ub lb country), xla(0/1, valuelabel noticks)  ///
        legend(label(1 "Estimates") label(2 "95% CI")) ///
        xscale(r(-3 4)) xsize(4)

[Matthew Williams]

Thank you, Andrew and Øyvind

Thank you so much for your quick help. It works perfectly. I have an additional question regarding your code, that how do you know that xscale should range from -2 to 4 (or -3 to 4) in my case?

[Andrew Musau]

I should have perhaps specified -2 3 range as 0 1 is exactly in the middle of this.

-2 -1 0 1 2 3

It's mental mathematics, but not completely symmetrical in this case!