Search All of the Math Forum:

Views expressed in these public forums are not endorsed by NCTM or The Math Forum.

Notice: We are no longer accepting new posts, but the forums will continue to be readable.

Topic: Curvefitting
Replies: 2   Last Post: Oct 2, 1996 9:28 AM

 Messages: [ Previous | Next ]
 Koichi S Kunitake Posts: 1 Registered: 12/7/04
Curvefitting
Posted: Sep 30, 1996 3:38 PM

Hi, as a new user of Matlab, I have been faced with the following
problem. I am doing error propagation, and lets say that I have the following
situation:

1) I make a scatterplot of 20 points on a non-log, x-y plot.
2) I use c2=curvefit(x,y,1) to fit the least squares line
3) I have to use the slope of this line in further calculations, and would
like to know how to take the error of the slope into account. I know
that it is not simply corrcoef(x,y) but that I need something like:
The slope = 2345.6 +/- 43.3

How can I get a +/- like shown above? I looked through Traver's Tutorial, as
well as the one by Sigmon, and the MIT help docs, and couldn't figure out how
to do this. If someone could help, I would greatly appreciate it.

Koichi Kunitake
<kkunitak@mit.edu>

Date Subject Author
9/30/96 Koichi S Kunitake
10/1/96 Paige Miller