Re: st: question on graphing

[Duha Altindag <daltin2@tigers.lsu.edu>]

I'm curious:
Isn't  "the relationship between predicted logearnings and age holding
the other variables constant" same thing as the partial derivative of
log earnings with respect to age?
And so this is just a line with slope _b[age]?

On Mon, Apr 4, 2011 at 12:23 PM, Nick Cox <njcoxstata@gmail.com> wrote:
> No add-ons needed.
>
> sysuse auto
> gen hiprice = price > 6000
> reg mpg weight hiprice foreign
> predict mypredict
> egen group = group(hiprice foreign), label
> separate mypredict, by(group) veryshortlabel
> scatter mpg weight || line mypredict? weight, sort c(L ..)
>
> Nick
>
> On Mon, Apr 4, 2011 at 6:10 PM, Dimitriy V. Masterov <dvmaster@gmail.com> wrote:
>> Susan,
>>
>> I think you can accomplish this with:
>>
>> (1) -margins- to get the profiles. Use the at option for the age variable.
>> (2) -parmest- or -stastby- to save the profiles as a dataset
>> (3) Plot the results using -eclplot-
>>
>> DVM
>>
>>
>> On Mon, Apr 4, 2011 at 12:54 PM, Averett, Susan L
>> <averetts@lafayette.edu> wrote:
>>> Hi, I am running a regression with logearnings as the dependent variable and age in  years, a female dummy variable and a dummy equal to 1 if the individual has a college degree. I would like to graph my age/earnings profile in Stata...i.e. I'd like to graph the relationship between predicted logearnings and age holding the other variables constant but I cannot figure out how to do this. I'd also then like to plot the lines separately for men and women on the same graph.
>>> Any suggestions?
>>>
>>> Here is my regression:
>>>
>>> regress lahe age female bachelor
>>>
>
> *
> *   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/
>

*
*   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/