Re: Re: st: Polynomial Fitting and RD Design

[Maarten Buis <maartenlbuis@gmail.com>]

On Fri, Sep 2, 2011 at 8:50 PM, Austin Nichols wrote:
>  I agree that global
> polynomials are generically a bad idea for the RD design, but
> "considering a fractional polynomial model instead," one should be
> comparing to the standard approach, not a polynomial.  I.e. Maarten's
> example graph does not provide evidence for a fractional polynomial
> model over local linear regression.

That is as I intended my example: a comparison of a global polynomial
with a fractional polynomial. The polynomials suggested by the
original poster looked suspicious to me, and I was trying to come up
with some alternative that was easy to implement and I overlooked that
all this was already better implemented in the -rd- package. I happily
bow to suprerior wisedom.

Hope this helps,
Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany


http://www.maartenbuis.nl
--------------------------

*
*   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/