You are emitting weird hex-stuff from your mail client. And from what
I can tell that is not a reproducible example even if hex-ation
problem gets fixed. There is a worked example in segments.
On Oct 9, 2009, at 10:45 AM, Steve Murray wrote:
<cb247343-29ec-495a-ba8e-d5a466bab...@comcast.net>
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Thanks for flagging up the 'segments' command. However=2C I'm having
troubl=
e getting it to work - this is probably due to me misunderstanding
the docu=
mentation for this command.
The plot section of my script appears as follows:
test3 <- 36:45
plot(data_means=2C type=3D"b"=2C pch=3D4=2C ylab=3D""=2C
xlab=3D""=2C xaxt=
=3D"n"=2C yaxt=3D"n"=2C col=3D3=2C ylim=3Dc(250=2C380))
abline(lm(data_means[36:45] ~ test3)=2C lty=3D2)
As you can see=2C
I cannot "see" anything. If you are under the impression that there
should have been an attachment, then (re-?)read the Posting Guide on
that topic.
I'm only plotting points 36 to 45 from the object 'data_m=
eans'. This intentionally results in much white space along the x-
axis unti=
l point 36 is reached. However=2C when inserting the trendline=2C
this runs=
along the entire length of the x-axis instead of just through points
36 to=
45. Ideally=2C a line which slightly overshoots the data subset
would look=
the best=2C but one constrained to the extents of the 36:45 would
also do =
the job just fine.
So=2C my question is=2C how do I use 'segments' (or otherwise) to
create a =
linear trendline which only extends through points 36:45? (and
overshoots a=
t either end very slightly=2C if possible=2C rather than running
along the =
entire length of the x-axis).
The basic strategy would be to calculate the x and y coordinates of
the two points at the end of the desired segment and then "say"
segments( x[1], y[1], x[2], y[2] )
Many thanks again=2C
Steve
----------------------------------------
CC: r-help@r-project.org
From: dwinsem...@comcast.net
To: smurray...@hotmail.com
Subject: Re: [R] Trendline for a subset of data
Date: Fri=2C 9 Oct 2009 09:27:43 -0400
On Oct 9=2C 2009=2C at 5:50 AM=2C Steve Murray wrote:
Dear all=2C
I am using abline(lm ...) to insert a linear trendline through a
portion of my data (e.g. dataset[=2C36:45]). However=2C I am finding
that whilst the trendline is correctly displayed and representative
of the data portion I've chosen=2C the line continues to run beyond
this data segment and continues until it intersects the vertical
axes at each side of the plot.
How do I display the line so that it only runs between point 36 and
45 (as shown in the example above) as doesn't continue to display a
line throughout the rest of the plot space?
?segments
=0A=
_________________________________________________________________=0A=
View your other email accounts from your Hotmail inbox. Add them
now.=0A=
http://clk.atdmt.com/UKM/go/167688463/direct/01/=
David Winsemius, MD
Heritage Laboratories
West Hartford, CT
______________________________________________
R-help@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.
