Re: st: Graphing longitudinal profiles

[Thomas Speidel <thomas@tmbx.com>]

Thank you Maarten, that solved my problem.

Thomas

On Thu, 1 Dec 2011 17:40:34 +0100, Maarten Buis wrote:

> On Thu, Dec 1, 2011 at 5:26 PM, Thomas Speidel wrote:
>
> > I am having difficulties creating a connected line graph for a
> > longitudinal dataset.Notice the vertical line going from 13 to 19 on
> > the y-axis. Why?
>
> The problem are the missing values. The conect(L) starts a new line 
> if
> t[_n+1] < t[_n]. So for id 88 to id 93 there will be a line going 
> from
> 19 to 14 at t=0, which is not what you want. One solution is in the
> example below:
>
> *------------ begin example ------------
> drop _all
> input id time value
> 84 0 15
> 84 1 18
> 84 2 17
> 88 0 19
> 88 1 .
> 88 2 .
> 93 0 14
> 93 1 19
> 93 2 19
> 94 0 16
> 94 1 20
> 94 2 20
> 97 0 13
> 97 1 .
> 97 2 .
> 98 0 17
> 98 1 19
> 98 2 20
> 99 0 17
> 99 1 18
> 99 2 20
> end
> sort id time
> gen byte notmiss = !missing(value)
> bys id : egen Nnotmist = total(notmiss)
> twoway line value time if Nnotmist >=2 , connect(L)
> *---------------- end example -----------------
>
> Hope this helps,
> Maarten
>
> --------------------------
> Maarten L. Buis
> Institut fuer Soziologie
> Universitaet Tuebingen
> Wilhelmstrasse 36
> 72074 Tuebingen
> Germany
>
> http://www.maartenbuis.nl [2]
> --------------------------
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search [3]
> * http://www.stata.com/support/statalist/faq [4]
> * http://www.ats.ucla.edu/stat/stata/ [5]



Links:
------
[1] mailto:thomas@tmbx.com
[2] http://www.maartenbuis.nl
[3] http://www.stata.com/help.cgi?search
[4] http://www.stata.com/support/statalist/faq
[5] http://www.ats.ucla.edu/stat/stata/

--
Thomas Speidel
*
*   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/