| Author |
 Topic  |
|
|
hernan354
Chile
2 Posts |
 Posted - 09/28/2011 : 05:10:43 AM
|
Origin Ver. and Service Release (Select Help-->About Origin):
Operating System:
Dear Oringin Users:
I am trying to fit a curve with a biexponential decay (y = A1*exp(-x/t1) + A2*exp(-x/t2) + y0 ) but is very importan for my star the fitting x=0 y must be equal to 1 if is it possible do that with origin? how can a do it?
Thank you very much
Hernan |
|
|
Hideo Fujii
USA
1582 Posts |
Posted - 09/28/2011 : 10:41:01 AM
|
Hi Hernan,
With your condition, y=1 at x=0, you can obtain a linear constraint:
A1+A2+y0=1;
So, in NLFit tool, choose the "Constraint" page in "Code" tab.
Then, you can turn ON the "Enable Linear Constraints" check box, and enter the above equation. The fit result should satisfy this constraint.
--Hideo Fujii
OriginLab |
Edited by - Hideo Fujii on 09/28/2011 10:42:03 AM |
 |
|
|
hernan354
Chile
2 Posts |
Posted - 09/28/2011 : 3:47:01 PM
|
Hello,
Thanks for your advice, but I didn't get that I want. This was my result. Do you know what is wrong?
A1+A2+y0=1
Equation y = A1*exp(-x/t1) + A2*exp(-x/t2) + y0
Adj. R-Square 0,9986
Value Standard Error
B y0 2,13456E-4 3,81735E-6
B A1 0,33624 6,33896E-4
B t1 10,16847 0,02953
B A2 0,5677 6,79918E-4
B t2 47,19833 0,03806
|
|
|
|
easwar
USA
1965 Posts |
Posted - 09/29/2011 : 10:36:57 AM
|
Hi,
Did you check if your fitting session converged? Best to send your data to tech support so they can check. You can use the "Send File to Tech support" link on top right of this page, and then refer to this forum post.
Easwar
OriginLab |
|
|
| |
Topic |
|
|
|
Biexponential function with restriction