Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: Sub-optimal smoothing behavior by stcurve?


From   "Dupont, William" <william.dupont@Vanderbilt.Edu>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: Sub-optimal smoothing behavior by stcurve?
Date   Fri, 18 Nov 2005 14:03:17 -0600

Yulia

Thanks for your explanation and suggestion.  I hope that you will decide
to implement boundary kernels as they will be very useful for many
users.  Limiting the plot to a range where it is valid is also a good
idea.  In the interim, you might also modify your documentation to warn
users of this boundary bias.  A problem with the current implementation
is the downward hook at the right end of the estimated hazard curve.
When time denotes age, or is correlated with age, it can be confusing as
to whether this represents a healthy survivor effect or is merely an
artifact of the smoothing algorithm.

Bill

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Yulia
Marchenko, StataCorp
Sent: Friday, November 18, 2005 1:23 PM
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: Sub-optimal smoothing behavior by stcurve?

William Dupont <william.dupont@Vanderbilt.Edu> asks about behaviour of
-stcurve- at the boundaries:

>I believe that the smoothing behavior for the hazard function plots of
>-stcurve- is less than ideal near the time boundaries.
>...
>I have not attempted to read the -stcurve- code and realize that 
>devising smoothing algorithms can be non-trivial.  I wondered, however,

>if the program was really working as the authors intended or it there 
>might be some way of improving its performance near the time
boundaries.

The algorithm in -stcurve- uses the usual smoothing kernel technique to
estimate hazard function as described in [ST] sts graph on p.292. Due to
the symmetry of the kernel, kernel estimators encounter bias at the
boundary points.

Two solutions to this problem would be

   1. Retrict the plot region to not include points near the boundary.
      This is something Bill can do himself, and something we are
considering
      doing officially.

   2. Use boundary kernels to alleviate the bias.  This is also
something
      we are considering doing officially.

-- Yulia
   ymarchenko@stata.com

*
*   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