st: RE: RE: AW: xtline, xtline2 and cmissing(n)

["Nick Cox" <n.j.cox@durham.ac.uk>]

I don't know about -xtline2-. But -linkplot- from SSC might offer you an
alternative. 

-linkplot- is not automatically geared to panel data, but linking on
panel identifier and plotting against time should produce the kind of
results I think you want. 

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

Giovanni Vecchi

thank you.

Your examples work fine.

However, I've just realized that a conflict arises if you also include
the
option -overlay- among the options (no matter whether you use -xtline-
or
-xtline2-).

Once I include both -cmissing(n)- and -overlay- ... I'm back to square
one:

. xtline2 x, overlay cmissing(n) name(mygr2, replace) nodraw

option cmissing() not allowed
r(198);

Martin Weiss
 
> In 10.1, I get the old behavior with -xtline-, and the new behavior,
i.e.
> respect for -cmissing()-, for -xtline2-
> (http://www.stata.com/statalist/archive/2009-10/msg00291.html), just
as
> promised:
> 
> 
> *************
> clear*
> vers 10.1
> 
> set obs 4
> gen id=_n
> expand 10
> bys id: gen time=_n
> gen x=runiform()
> xtset id time
> replace x=. if runiform()<.3
> 
> 
> xtline x, cmissing(n) name(mygr, replace) nodraw
> xtline2 x, cmissing(n) name(mygr2, replace) nodraw
> 
> graph combine mygr mygr2, col(1)
> *************
> 
> 
> In 11, both yield identical results.
 
Giovanni Vecchi
 
> following up on a couple of emails on the failure of -xtline- to
account
> for
> the -cmissing(n)- option
> (http://www.stata.com/statalist/archive/2009-10/msg00291.html), I'm
writing
> to report that, in my experience, the bug in -xtline- has not been
fixed.
> Nor does -xtline2-, kindly provided by Jeff Pitblado, work for me.
> 
> After running a number of experiments under both Stata 10.1 and Stata
11,
> I'm now short of ideas and seeking for your advice.

*
*   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/