| T O P I C R E V I E W |
| erickzhou |
Posted - 07/15/2010 : 08:56:29 AM
Origin Ver8. and Service Release (Select Help-->About Origin):
Operating System:winXP
In Origin 8, there is one menu of "nonlinear surface fit". If I want to get the 2-D polynominal trend of my data, is it appropriate to choose this menu. My data is composed by a two dimension polynominal trend fit and gaussion noise.
There is another menu of "multiple regresssion", I don't know whether this menu is better for my aim of trend extraction. Is there any difference between "nonlinear surface fit" and "multiple regresssion", or maybe the "nonlinear surface fit" in Origin 8 is just a kind of "multiple regresssion"? |
| 3 L A T E S T R E P L I E S (Newest First) |
| erickzhou |
Posted - 07/16/2010 : 10:22:50 AM
Thank you! |
| larry_lan |
Posted - 07/15/2010 : 10:27:20 PM
Hi:
Nonlinear surface fitting fits nonlinear model with two independent parameters; Multiple Regression fits linear model. You should choose proper method according your model and research goal.
If the model is linear, nonlinear fitting will generate the same fitted parameter values with multiple regression.
Thanks
Larry |
| erickzhou |
Posted - 07/15/2010 : 9:43:24 PM
I am sorry for a mistake. The word "polynominal" should be "polynomial".
My data is composed by a two dimension second-order polynomial trend and gaussion noise. If I want to extract the 2-D
second-order polynomial trend. what menu in Origin 8 should I choose, "nonlinear surface fit" or "multiple regresssion"? |
