Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: pspline help
From
"Manisha Chakrabarty" <[email protected]>
To
[email protected]
Subject
st: pspline help
Date
Sun, 15 Dec 2013 19:38:11 +0100
Dear Roberto G. Gutierrez
In your pspline , do you have any technical note for that?
for example how to choose knots and positins of it.Or do
you have any example. do file?
In my case I have to estimate semiparametric model using
pspline with some covariates.
Now while it does it for all observations in your dats
set, if for example I wish to incorporate the location
information of observational units (i.e, for each
observation i take all those neighbouring observation
depending on whether they share a common boundary of
district), then the knot positions will change, right?
This check has to be done for all observations i and
then we get final estimate of nonparametric part as a
spline as well as coefficients of parametric part. so if
i=1,..n, we will get out final result for n observations,
but pspline should be done by considering the neighborhood
observsations for ecah i.
what should I do in that case?
Best
Manisha
*
* 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/