# RE: st: AW: simple "graph twoway, b()" question

 From "Nick Cox" To Subject RE: st: AW: simple "graph twoway, b()" question Date Wed, 25 Feb 2009 17:30:30 -0000

```You may get better control by specify -xla()- instead.

Nick
n.j.cox@durham.ac.uk

Martin Weiss

Well,  how close is the third line in my suggestion

line le year, xsc(r(1850 2006)) by(x)

Markus Pannenberg

Thanks Martin, you are right! (I also figured this out before sending
my e-mail to the stata-list....).

However: Does anyone know a work-around...?

On Wed, Feb 25, 2009 at 5:53 PM, Martin Weiss <martin.weiss1@gmx.de> wrote:
>
> The thing - in my view - is that not every -twoway- option carries over to
> the -by- options. See

> help by_option

Try

sysuse uslifeexp, clear
set seed 1001
g x =runiform()>0.5
line le year, by(x)
line le year, xsc(r(1850 2006))
line le year, xsc(r(1850 2006)) by(x)
line le year, by(x, xsc(r(1850 2006)))

and note the different behavior.

Markus Pannenberg

> I have a simple "graph twoway, by()" question:
>
> When I type:
>
> (1) twoway (line gdp year),  by(country),
>
> Stata creates a graph with plots for all countries and a range for the
> x axis starting in 2000 though the minimum of the variable year is
> 2001!
>
>
> When I type:
>
> (2) twoway (line gdp year),  xsc(r(2001 2006)),
>
> Stata creates a graph with a range for the x axis starting in 2001.
>
> However, when I combine (1) and (2), i.e.
>
> (3) twoway (line gdp year), by(country, xsc(r(2001 2006)))
>
> Stata tells me: "option xsc() not allowed" r(198)!

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
```