Path: news.mathworks.com!not-for-mail
From: "Daniel Robbins" <d.w.e.robbins@gmail.com>
Newsgroups: comp.soft-sys.matlab
Subject: Identifying linear vs  curvilinear relationship
Date: Fri, 8 Mar 2013 11:45:07 +0000 (UTC)
Organization: The MathWorks, Inc.
Lines: 11
Message-ID: <khcj02$5id$1@newscl01ah.mathworks.com>
Reply-To: "Daniel Robbins" <d.w.e.robbins@gmail.com>
NNTP-Posting-Host: www-06-blr.mathworks.com
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
X-Trace: newscl01ah.mathworks.com 1362743107 5709 172.30.248.38 (8 Mar 2013 11:45:07 GMT)
X-Complaints-To: news@mathworks.com
NNTP-Posting-Date: Fri, 8 Mar 2013 11:45:07 +0000 (UTC)
X-Newsreader: MATLAB Central Newsreader 2776992
Xref: news.mathworks.com comp.soft-sys.matlab:790639

Hi,

Can anyone advise on the best method to identify if the relationship between two varibales is linear or curvilinear? I am collecting data which includes both.

Potentially I could calculate both and then use an 'IF' function to detect the highest R^2 value, but this seems long winded. I am presuming there is a better approach to automate this process.

FYI, I dont have the curvefitting toolbox...

Any thoughts are greatly appreciated!
Regards
Dan