[R] Odp: connecting points into a smooth curve

[Petr PIKAL]

Hi

r-help-bounces at r-project.org napsal dne 01.11.2010 13:02:46:

> 
> 
> 
> 
> 
> 
> ----------------------------------------
> > To: kaiyan at berkeley.edu
> > From: petr.pikal at precheza.cz
> > Date: Mon, 1 Nov 2010 11:50:17 +0100
> > CC: r-help at r-project.org
> > Subject: [R] Odp: connecting points into a smooth curve
> >
> > Hi
> >
> > r-help-bounces at r-project.org napsal dne 01.11.2010 07:18:47:
> >
> > >
> > > If I have, say, five scatter points and want to connect them 
together
> > into a
> > > smooth curve.
> >
> > As you are not much specific about what you consider "smooth curve" 
here
> > are some options
> 
> 
> I think the OP said he had 5 points so that leaves a lot to the 
imagination.
> Presumably smooth means something like, " a curve that goes through all 
5 points
> and has as many continuous derivatives as possible." So you write a 
general
> fitting function, Taylor would be a good expansion I suppose, impose
> the above conditions, and see if there
> is an R package that provides the coefficients you want? 
> Someone else was asking a related question and there I offered sinc 
interpolation
> and indeed this can be exact given a band limited signal and no time 
base errors etc.
> 
> 
> Doing much of anything meaningful with 5 points would probably require a 
model
> as the other
> poster suggested- your model would need to be solved depending on its 
particulars.
> 
> You sometimes see these kinds of wild interpolation issues with the 
drawing 
> programs and free-form input "smoothing" where it tries to fit a smooth 
curve 
> to your mouse moves. 

And of course in Excel graphs you can find something like "smooth" 
connection to points. Therefore the method depends on what the OP means by 
smooth curve.

Regards
Petr




> 
> 
> >
> > use some model ?lm and plot predicted values with line
> > ??smooth will give you many functions for smoothing data e.g.
> > ?loess
> > ?supsmu
> > ?spline
> > and some others
> >
> > Regards
> > Petr
> >
> > > I did plot(x,y,type="l"), but the graph is five segments connecting 
with
> > > each other, but not a smooth curve.
> [[elided Hotmail spam]]
> > >
> > >
> > > --
> > > View this message in context:
> > http://r.789695.n4.nabble.com/connecting-points-
> > > into-a-smooth-curve-tp3021796p3021796.html
> > > Sent from the R help mailing list archive at Nabble.com.
> > >
> > > ______________________________________________
> > > R-help at r-project.org mailing list
> > > https://stat.ethz.ch/mailman/listinfo/r-help
> > > PLEASE do read the posting guide
> > http://www.R-project.org/posting-guide.html
> > > and provide commented, minimal, self-contained, reproducible code.
> >
> > ______________________________________________
> > R-help at r-project.org mailing list
> > https://stat.ethz.ch/mailman/listinfo/r-help
> > PLEASE do read the posting guide 
http://www.R-project.org/posting-guide.html
> > and provide commented, minimal, self-contained, reproducible code.
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide 
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.