Mime-Version: |
1.0 |
Sender: |
|
Subject: |
|
From: |
|
Date: |
Wed, 20 Jul 2005 16:02:46 -0400 |
Content-Transfer-Encoding: |
8bit |
Content-Type: |
text/plain; charset=US-ASCII |
Reply-To: |
|
Parts/Attachments: |
|
|
Normally, what I've seen people do is that, if you have enough data, you take 1/3 of the data to worm up the model, then the rest of the data, I would divide in half for calibration and validation.
Ali
Ali M. Sadeghi, Soil Physicist
Environmental Quality Laboratory
USDA-ARS, BARC-West,B-007,R-211
10300 Baltimore Ave.
Beltsville, MD 20705
Phone:(301) 504-6693
Fax: (301) 504-5048
Email: [log in to unmask]
http://www.barc.usda.gov/anri/eql/sadeghi.html
>>> [log in to unmask] 07/20/05 03:01PM >>>
Dear allI am searching for information related with model calibration and validation. My specific doubts are partitioning of a given data set for calibration and validation purposes. For example if I have 20 data how many should I use for calibration? Does any body know of rules for this purpose?Thanks,Juan Pablo Monzon, Argentina
|
|
|