[R] dnorm Line Plot Graphics

[(Ted Harding)]

On 13-Jul-10 22:21:29, Melissa Peters wrote:
> Hello,
> 
> 1. I have a number of snow depth measurements (centimeters depth) that
> were acquired in a row (linear transect). 
> 
> A: I am trying to plot the probability density function (PDF) of these
> points. When I use the "dnorm" command and line type = points, the PDF
> is displayed correctly. When I change the line type = line, it displays
> the PDF curve with numerous lines connecting multiple of the points
> together rather than just one line connecting the points in sequence. I
> have noticed in some online help tutorials the use of the "seq"
> command, but when I employ this it simply flattens my PDF curve to a
> straight line at 0% probability. Is there something I am missing?
> 
>> plot(Snow,dnorm(Snow,mean=26.61,sd=14.179),type="l")

On just this point: From your description, it seems likely that the
values in "Snow" are not in increasing order. Hence, when you plot
usinf type="l", ir will take the points in the order in which they
are found in "Snow" and so the lines connecting successive points
will jump about all over the place. You can cure this by sorting
"Snow" into increasing order:

  plot(sort(Snow),dnorm(sort(Snow),mean=26.61,sd=14.179),type="l")

should do it. However, this may not look as nice as you would like,
if the spacings between successive sorted values are very irregular.
It may look better if you plot "Snow" as points, and then add
lines drawn to a regular sequence. Hence something like

  S0 <- ((max(Snow) - min(Snow))/100)*(-0.05+(0:101))
  plot(Snow,dnorm(Snow,mean=26.61,sd=14.179),pch="+",col="blue")
  lines(S0,dnorm(S0,mean=26.61,sd=14.179),col="green")

may look better.

Hoping this helps,
Ted.

--------------------------------------------------------------------
E-Mail: (Ted Harding) <Ted.Harding at manchester.ac.uk>
Fax-to-email: +44 (0)870 094 0861
Date: 14-Jul-10                                       Time: 00:08:32
------------------------------ XFMail ------------------------------