Dear all,
I am currently estimating more-way fixed effects regressions using
Thomas Cornelissens felsdvreg-command (Cornelißen, T. [2008]: The
stata module felsdvreg to estimate a linear model with two
high-dimensional fixed effects, The Stata Journal, 8(2), pp. 170-189.)
using linear splines in x. After estimation I want to graph the
relationship between the outcome and x (which is not the problem) and
add confidence intervals that reflect the fact that not all of the
estimates parameters are significant (similar to -lpoly y x, ci-).
Unfortunately, I can't get the last part to work.
I have created a small example using the NLSY-data used in [XT] and
simple FE-regressions to illustrate the point.
use http://www.stata-press.com/data/r10/nlswork, clear
xtset idcode year
mkspline agespl 20 = age, dis
xtreg ln_wage agespl*, fe cluster(idcode)
predict yhat
twoway line yhat age, sort
lpoly ln_wage age, ci noscatter
What I would like to do is adding confidence intervals to to the first
graph that reflect the fact that, e.g., agespl1 is estimated with less
precision than, e.g., agespl3 - similar to the -lpoly- results. Is
there any way to do this?
Any help would be greatly appreciated.
Best regards,
Nils
*
* 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/
