[R] derivatives from loess (not locpoly)?
andy_liaw at merck.com
Tue May 20 06:56:32 CEST 2003
A possible alternative is locfit(..., deriv=...) (in package `locfit'), as
described in Section 6.1 of Loader's "Local Regression and Likelihood".
> -----Original Message-----
> From: Prof Brian Ripley [mailto:ripley at stats.ox.ac.uk]
> Sent: Sunday, May 18, 2003 3:12 AM
> To: J S
> Cc: r-help at stat.math.ethz.ch
> Subject: Re: [R] derivatives from loess (not locpoly)?
> Not easily. Unlike locpoly (in KernSmooth, uncredited), loess uses a
> locally varying bandwidth and that would need to be taken
> into account in
> determining the derivative. The underlying Fortran code does
> not have any
> support for this.
> On Sat, 17 May 2003, J S wrote:
> > is there a way of estimating derivative curves, similar to
> the ones we
> > get from 'locpoly', from 'loess' estimation.
> > i am interested in estimation of 1st and 2nd derivatives...
> Brian D. Ripley, ripley at stats.ox.ac.uk
> Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
> University of Oxford, Tel: +44 1865 272861 (self)
> 1 South Parks Road, +44 1865 272866 (PA)
> Oxford OX1 3TG, UK Fax: +44 1865 272595
> R-help at stat.math.ethz.ch mailing list
More information about the R-help