Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: Re: Re: Re: Fracpoly


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Re: Re: Re: Fracpoly
Date   Tue, 12 Feb 2008 00:41:27 -0000

-ereturn list- shows what -fracpoly- leaves in memory. In principle, you
could work on that information. In practice, you would need to work up
some Stata skills to get there. Doing this in complete generality might
be quite an undertaking. 

I have some broad concern with what you are doing. -fracpoly- fits a
model to the whole dataset. It often works nicely because it uses quite
a flexible family of possible fits. Nevertheless you want to extract
properties that depend on first and second derivatives. That seems to me
to invest a great deal of trust in the models, which are not necessarily
optimal for that purpose. However, I don't have a good alternative
solution, but I would recommend some comparison with quite different
ways of doing it, e.g. based on numerical differentiation of other kinds
of smooth fit. 

Nick 
n.j.cox@durham.ac.uk 

mai mai

I guess I am wondering is there a way to enter the resulting
functional form that is predicted by Fracpoly into STATA
automatically. Unfortunatly, I am  not technically savvy with STATA to
figure out a way to do that.


*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index