Re: st: How to graph several mean trajectories and combine them in a single graph

[Alan Acock <acock@mac.com>]

Try it with the graph editor.

Alan
On Feb 25, 2014, at 2:19 PM, Han-Jung Ko (Koko) <kohan@onid.oregonstate.edu> wrote:

> Thank you both for the help.
> 
> I did try your suggestions and did more research. I am able to now
> graph the dots of means of my variables of interest by wave in a
> single graph (i.e., means of pr_ and pl_). Here is my command:
> 
> egen pl_m = mean(pl_), by (wave)
> egen pr_m = mean(pr_), by(wave)
> graph twoway dot pl_m pr_m wave
> 
> However, when I tried to connect the dots using the following command:
> graph twoway connected pl_m pr_m wave
> 
> it not only connects the dots but also draw a line from wave 1 dot to
> wave 5 dot for pr_ and pl_. Do you know how I can do to have only the
> dots connected for pl_ and pr_ separately but not a straight line from
> wave 1 to wave 5 datapoints?
> 
> Hope the question is clear. I am sorry that I am not a native speaker.
> 
> Sincerely,
> Han-Jung Ko
> 
> Han-Jung Ko, M.S.
> Doctoral Candidate
> Human Development and Family Studies
> School of Social and Behavioral Health Sciences
> Oregon State University
> kohan@onid.orst.edu
> 
> 
> On Tue, Feb 25, 2014 at 1:53 PM, Nick Cox <njcoxstata@gmail.com> wrote:
>> -profileplot- does require a -by()- option but feeding it a variable
>> that is constant suppresses subdivision when you don't want it.
>> 
>> Nick
>> njcoxstata@gmail.com
>> 
>> On 25 February 2014 18:55, Han-Jung Ko (Koko)
>> <kohan@onid.oregonstate.edu> wrote:
>>> Thank you. I am not certain. The profileplot command seems requiring
>>> by() specification. I may try to see whether graph command could work
>>> for graphing means across individuals for certain variables overtime
>>> and combine them...
>> 
>> On Tue, Feb 25, 2014 at 11:57 AM, Nick Cox <njcoxstata@gmail.com> wrote:
>> 
>>>> Sorry, but I can't advise further given "did not work". What was wrong
>>>> from your point of view?
>> 
>>>> On 25 February 2014 17:49, Han-Jung Ko (Koko)
>>>> <kohan@onid.oregonstate.edu> wrote:
>>>>> Thank you, Nick. I tried your suggestion and it did not work though.
>>>>> I will try to ask in the FAQ in the link you sent me.
>> 
>>>>> On Tue, Feb 25, 2014 at 3:31 AM, Nick Cox <njcoxstata@gmail.com> wrote:
>>>>>> -profileplot- is a user-written program by Philip Ender and from
>>>>>> http://www.ats.ucla.edu/stat/stata/ado/analysis
>>>>>> as you are asked to explain in the FAQ.
>>>>>> 
>>>>>> If I understand you correctly, you can easily fool -profileplot- into
>>>>>> lumping all observations together by
>>>>>> 
>>>>>> gen everyone = 1
>>>>>> 
>>>>>> profileplot ..., by(everyone)
>>>>>> 
>>>>>> although you will probably want to fix legend.
>>>>>> 
>>>>>> Nick
>>>>>> njcoxstata@gmail.com
>>>>>> 
>>>>>> On 24 February 2014 20:44, Han-Jung Ko
>>>>>> 
>>>>>>> I am trying to graph several mean trajectories and combine them in a
>>>>>>> single graph. By using profileplot command, it is possible to graph
>>>>>>> them side-by-side but not in a single graph.
>>>>>>> For example,
>>>>>>> . profileplot pl_1 pl_2 pl_3 pl_4 pl_5, by(sex) name(graph_pl, replace)
>>>>>>> . profileplot pg_1 pg_2 pg_3 pg_4 pg_5, by(sex) name(graph_pg, replace)
>>>>>>> . graph combine graph_pl graph_pg
>>>>>>> 
>>>>>>> However, using profileplot command also requires the option by(). I
>>>>>>> wonder how I could graph the meantrajectories for all participants
>>>>>>> (not by gender) and combine the total of six trajectories together?
>> *
>> *   For searches and help try:
>> *   http://www.stata.com/help.cgi?search
>> *   http://www.stata.com/support/faqs/resources/statalist-faq/
>> *   http://www.ats.ucla.edu/stat/stata/
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/

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