Coefplot by year

[Rui Tiago]

Dear Stata users,
I am estimating the following regressions:
reg wage age agesq i.educ if year==2000
est sto m1
reg wage age agesq i.educ if year==2001
est sto m2
reg wage age agesq i.educ if year==2002
est sto m3
I want to create a graph that shows the evolution of the standard deviation of wages, meaning I want the year in the x axis and the standard deviation of wage in the y axis. Could anyone explain me how I can do this? It seems that I need to use the command coefplot, but I do not know how to continue. Is it also possible to use instead of the "standard deviation of wage" the "standard deviation of wage residuals"?

[ericmelse]

Dear Rui,

Extensive documentation can be found here and here, a working paper here, a TSJ paper here, a workshop presentation here (but not answering your question), a nice blog about CI's here, and Ben Jann's (the coefplot author) German presentation here.
But the first step, of course, is to install coefplot and consult the helpfile:

ssc install coefplot
adoupdate coefplot
h coefplot

Best,
Eric

[Rui Tiago]

Thank you, Eric Melse for your help.
I tried to improve my code looking to the documentation:
forv i=2000/2003 {
quietly reg lghwage age agesq if year==`i'
est sto `i'
}
coefplot 2000 || 2001 || 2002 || 2003, drop(age agesq) bycoefs byopts(yrescale) vertical

But I receive the message 2000 invalid name r(7).
I also was not able to understand how can I create a continuous line.