s-news
[Top] [All Lists]

[S] loess vs lowess: please forget previous question

To: "'s-news@wubios.wustl.edu'" <s-news@wubios.wustl.edu>
Subject: [S] loess vs lowess: please forget previous question
From: "Laffont, Jean-Louis" <Laffont@phibred.com>
Date: Fri, 26 Jun 1998 06:47:54 -0500
Cc: "Laffont, Jean-Louis" <Laffont@phibred.com>
Sender: owner-s-news@wubios.wustl.edu
Dear S+ users,

Sorry: I have realized that I was missing the degree parameter in loess
which has value 2 by default. Setting degree=1 provides same results
than lowess.
Sorry again,
Jean-Louis

-----------------------
Original message:

> Dear S+ users,
> 
> I have been using the lowess function to add a smooth curve to my
> scatter plots.
> Now, I want to use the loess function but I am having trouble because
> I am getting different results with the two functions. Here is an
> example:
> 
>  attach(ethanol)
>  ethanol.loess<-loess(NOx~E,degree=1,span=2/3)
>  
> plot(ethanol.loess,xlim=range(E),ylim=range(NOx,fitted(ethanol.loess))
> )
>  points(E,NOx)
>  lines(lowess(E,NOx,f=2/3),col=2)
> 
> The two curves are very different (visually, I prefer the one produced
> with loess).
> Am I leaving something out in setting up the parameters?
> 
> Thanks for your help.
> 
> Jean-Louis
> (I am using S+ 4.5 Professional, release 1)
> 
> ---------------------
> Jean-Louis Laffont
> Pioneer Semences
> Aussonne - France
> email: laffont@phibred.com
> 
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu.  To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message:  unsubscribe s-news

<Prev in Thread] Current Thread [Next in Thread>
  • [S] loess vs lowess: please forget previous question, Laffont, Jean-Louis <=