Message-ID: <Pine.GSO.4.58.0408011210260.26773@panther.cs.ucla.edu>
Date: 2004-08-01T19:12:20Z
From: Yan Yu
Subject: locfit
In-Reply-To: <Pine.LNX.4.44.0408010826560.16672-100000@gannet.stats>
Got it, Thanks a LOT for the info!
yan
On Sun, 1 Aug 2004, Prof Brian Ripley wrote:
> The short answer is `no', assuming you are talking about the function in
> the CRAN package called `locfit' (you did not say).
yes, that is "locfit" package i am taking about, sorry that i forgot to
mention that..
> locfit has many options, but all do local smoothing, not interpolation.
>
> On Sat, 31 Jul 2004, Yan Yu wrote:
>
> > Hi, I have a Q on locfit,
> > E.g., I have an input data set of the form {(x, y, v)} , (x, y) is the
> > location in a 2D space, v is the value at that location.
> > I am wondering for the output of locfit(), if a given point (x, y) is in
> > the input data set, is the value at (x, y) going to be the exactly same
> > as the input? or it depends on the local fitting function?
> > does my Q make sense?
>
>